Complex media supplemented with a carbon source are commonly used in bioprocesses for recombinant protein production in E. coli. Optimizing these processes is challenging and requires precise understanding of cellular metabolism and nutrient requirements. Compared to a Design of Experiments approach that necessitates extensive experimentation, metabolic modeling using a genome scale metabolic model (GEM) offers a more predictive and systematic approach to guide process optimization by identifying specific metabolic bottlenecks. In addition, spent media analysis (SMA) can unravel the preferential utilization of different media components during the bioprocess. Here, we integrated the updated E. coli GEM with time course SMA data from a fed-batch process and performed dynamic flux balance analysis (dFBA) to identify metabolites that function as active nodes and are vital for cellular function. These are potential target supplements to boost cellular activity and in turn the recombinant protein productivity. Using an iterative approach of performing fermentation, SMA, and metabolic modeling, we optimized the bioprocess in just five experimental trials, resulting in a six-fold increase in protein productivity. The optimized feeding strategy involved yeast extract with amino acid supplementation (Ser, Thr, Asp, Glu) and increased oxygen transfer rates. This approach demonstrates significant promise for application in bioprocess optimization.

Multispecific antibodies are increasingly being explored in the pharmaceutical industry for unmet patient needs. This work focuses on generating these molecules through an electrostatic-steering strategy, where two separate parent homodimer antibodies are expressed and purified, then combined into the heterodimer multispecific through reduction and oxidation chemistry. Traditional operations for electrostatic steering multispecifics can include complex processing steps. Therefore, a novel redox process to generate the multispecific has been explored. This process involves a column-based reduction reaction and a spike of oxidant in the elution pool to form the heterodimer. This new strategy can simplify the downstream purification process for electrostatic-steering based molecules. The method consists of simultaneously binding two separate parental homodimers to the protein A chromatography resin and applying a reductant wash to reduce the interchain disulfide bonds. The molecules are then eluted, neutralized, and oxidized to form the intact heterodimer. The mechanism and rates of reduction, heterodimerization, and oxidation have been characterized to maximize conversion and product quality. This strategy has been demonstrated successfully for five multispecifics with diverse specificity and IgG subclasses. Implementing this method for pharmaceutical bioprocesses in the production of multispecific molecules offers the potential for the reduction in manufacturing complexity while maintaining acceptable product quality and yield.

Multispecifics are increasingly being evaluated in the pharmaceutical industry due to the unique mechanisms of action of the molecules, which is enabled by the multiple antigen binding capability. The complexity of these molecules can make production difficult, therefore different approaches for the generation of these molecules have been developed. The approach employed in this study utilizes electrostatic-steering, wherein charge-based differences between two parental homodimer antibodies are used to drive correct heterodimerization during a redox reaction of the partially purified parental homodimers. This strategy results in high conversion to the heterodimer with minimal product-related impurities; however, this method also requires separate bioreactors for each parental homodimer, resulting in complex manufacturing campaigns. This work describes a new bioprocess for electrostatic steering-based multispecifics. This strategy couples two unique components. First, the two separate cell lines are co-cultured, resulting in simultaneous production of both parental homodimers in a single bioreactor. The second component utilizes a column-based redox reaction, wherein the homodimers are captured and the disulfide bonds reduced while bound to the protein A resin using a reductant wash. The column is then eluted and neutralized to allow the reduced parental homodimers to heterodimerize and finally, the addition of an oxidant enables the disulfide bond reformation to complete the formation of the multispecific. This new process is robust and efficient across bench and manufacturing scales, with well-controlled impurity profiles. Through this strategy, the new multispecific bioprocess is more similar to a typical antibody-like bioprocess, enabling more efficient use of clinical and commercial manufacturing resources, while resulting in the production of complex multispecific molecules with minimal product-related impurities.

Plasmid DNA (pDNA) is a vital product for the pharmaceutical industry, which can be injected directly in the form of DNA vaccines, and its significance and demand are increasing. The DNA vaccines require high production of the pDNA in form of supercoiled plasmid DNA (sc-pDNA) for their actual development. Several factors, including the host strain, the type of plasmid, and production process, influence the production and productivity of pDNA. In this study, the rnaII gene, necessary for ColE I plasmid replication initiation, along with the nrdAB genes involved in the dNTP’s synthesis pathway, were integrated into the Escherichia coli ( E. coli) genome. This integration effectively improved plasmid production by enhancing the replication initiation and increasing the supply of replication substrates. Additionally, a rich fed fermentation medium was used, and the fermentation process was optimized. After an initial period of fermentation, a gradient-fed strategy was implemented to control specific growth rates. As a result, the fermentation broth reached an OD 600 of 113.7, with a plasmid production of 2.77 g/L and a supercoiled proportion of 97.3%.

In filamentous fungal fermentation, selecting suitable impellers and controlling fungal morphology are crucial for product yield. We previously demonstrated that the AGΔ-GAGΔ strain of Aspergillus oryzae, lacking both α-1,3-glucan (AG) and galactosaminogalactan (GAG), had improved hyphal dispersion, reduced culture viscosity, and increased recombinant protein production. Here, we applied computational fluid dynamics (CFD) using viscosity data to explore the importance of impeller design and strain engineering. High-performance impellers (HS100/HR100) were compared with the conventional flat-blade turbine and paddle (6FT/4FP). CFD analysis showed large gas cavities behind the blades and severe compartmentalization for both the wild-type and AGΔ-GAGΔ strains. HS100/HR100 had more homogeneous velocity and shear stress values than did 6FT/4FP. The AGΔ-GAGΔ strain had a wider shear stress distribution and reduced gas cavity in comparison with the wild-type strain. The simulation results agreed with measured volumetric oxygen mass transfer coefficients ( KLa) and mixing times. Transcriptional analysis during HS100/HR100 cultivation revealed upregulation of the TCA cycle genes in AGΔ-GAGΔ in comparison with the wild type. Our findings suggest that the combination of an optimized stirring system and the AGΔ-GAGΔ strain can significantly enhance mixing. Furthermore, the improved mixing properties of AGΔ-GAGΔ may contribute to higher recombinant protein yields through an increased population of metabolically active cells.

An immobilized, cell-free system to produce isobutanol, a next-generation biofuel, was engineered to improve its production using biochemical approaches. A previously described approach involving random immobilization on suspended beads was superseded by a targeted, ordered cellulosome-based immobilization system that facilitated substrate channeling and minimized diffusion limitations. For this purpose, keto-acid decarboxylase, alcohol dehydrogenase, and formate dehydrogenase were genetically fused to a specific dockerin module, originating from different microbial species, each of which bound selectively to a complementary cohesin-bearing scaffoldin protein. Engineering of the fusions, as well as tuning the expression, was key to obtaining a working efficient multi-enzyme system. This system produced isobutanol at a titer of 5.92 g/L and 78.4% yield, which represented a marked improvement over previous results that used a cell-free method to produce this biofuel. This approach is promising, both as a starting point for further expansion of biofuel production and as a general enzyme-immobilized platform for producing biochemicals in vitro.

Changes in biological pathways provide essential clues about metabolism. Genome-scale metabolic Models (GEM) are network-based templates that computationally describe all stoichiometric associations and gene-protein reaction (GPR) relations found in an organism for all its metabolic genes and metabolites. Using reaction stoichiometry as input, GEMs mathematically simulate metabolic reaction fluxes occurring in an organism and predict changes in the metabolic system under the relevant condition. Multiple tools and approaches in the literature can capture fluxes sensitive to a given condition by using GEMs. However, functional enrichment analysis of these reaction lists in a systems biology perspective is not straightforward. Here, we introduce RSEA to annotate given reaction sets to significantly related metabolic pathways: Reaction Set Enrichment Analysis web server tool. RSEA converts given reaction list derived from GEMs into proper reaction identifiers and statistically analyze its enrichment in metabolic pathways. RSEA is designed to provide researchers with a practical and user-friendly platform to explore and interpret sets of reactions in biological pathways and freely available online (https://rseatool.com/).

Increasing demand for recombinant adeno-associated virus (rAAV)-based gene therapies necessitates increased manufacturing production. Transient transfection of mammalian cells remains the most commonly used method to produce clinical-grade rAAVs due to its ease of implementation. However, transient transfection processes are often characterized by sub-optimal yields and low fractions of full-to-total capsids, both of which contribute to the high cost of goods of many rAAV-based gene therapies. Our previously developed mechanistic model for rAAV2/5 production indicated that the inadequate capsid filling is due to a temporal misalignment between viral DNA replication and capsid synthesis within the cells and the repression of later phase capsid formation by Rep proteins. We experimentally validated this prediction and showed that performing multiple, time-separated doses of plasmid increases the production of rAAV. In this study, we use the insights generated by our mechanistic model to develop an intensified process for rAAV production that combines continuous perfusion with high cell density re-transfection. We demonstrate that performing multiple, time-separated doses at high cell density boosts both cell-specific and volumetric productivity and improves plasmid utilization when compared to a single bolus at standard operating conditions. Our results establish a new paradigm for continuously manufacturing rAAV via transient transfection that improves productivity and reduces manufacturing costs.

Sub-models representing transformation processes by microorganisms and hydrolases, a one-dimensional (1-D) biofilm, and a bioreactor were integrated to simulate organic-matter fermentation and methane (CH 4) production in an anaerobic moving bed biofilm reactor (AnMBBR). The integrated models correctly represented all experimental observations and identified mechanisms underlying how and why AnMBBR performance changed when the volumetric loading rate (VLR) of total chemical oxygen demand (TCOD) increased from 3.9 to 19.5 kg COD T/m 3-d. The fractional removal of TCOD and CH 4 production decreased as the VLR of TCOD increased due in part to an increasing biofilm thickness that filled the interior channels of plastic carriers and led to a decrease in biofilm surface area and an increase in the mass-transfer boundary layer. Also, the ~25-day duration for each VLR of TCOD was too short to allow the biofilm to establish a new quasi steady state with respect to biofilm thickness. The mechanistic understanding of how biofilm characteristics and process performance respond to increased VLR of TCOD can be applied in engineering practice to improve AnMBBR process design and operation.

Recent advancements in three-dimensional (3D) cell culture technologies, such as cell spheroids, organoids, and 3D bioprinted tissue constructs, have significantly improved the physiological relevance of in vitro models. These models better mimic tissue structure and function, closely emulating in vivo characteristics and enhancing phenotypic analysis, critical for basic research and drug screening in personalized cancer therapy. Despite their potential, current 3D cell culture platforms face technical challenges, which include user unfriendliness in long-term dynamic cell culture, incompatibility with rapid cell encapsulation in biomimetic hydrogels, and low throughput for compound screening. To address these issues, we developed a 144-pillar plate with sidewalls and slits (144PillarPlate) and a complementary 144-perfusion plate with perfusion wells and reservoirs (144PerfusionPlate) for dynamic 3D cell culture and predictive compound screening. To accelerate biomimetic tissue formation, small Hep3B liver tumor spheroids suspended in alginate were printed and encapsulated on the 144PillarPlate rapidly by using microsolenoid valve-driven 3D bioprinting technology. The microarray bioprinting technology enabled precise and rapid loading of small spheroids in alginate on the pillar plate, facilitating reproducible and scalable formation of large tumor spheroids with minimal manual intervention. The bioprinted Hep3B spheroids on the 144PillarPlate were dynamically cultured in the 144PerfusionPlate and tested with anticancer drugs to measure drug effectiveness and determine the concentration required to inhibit 50% of the cell viability (IC 50 value). The perfusion plate enabled the convenient dynamic culture of tumor spheroids and facilitated the dynamic testing of anticancer drugs with increased sensitivity. It is envisioned that the integration of microarray bioprinting of tumor spheroids onto the pillar plate, along with dynamic 3D cell culture in the perfusion plate, could more accurately replicate tumor microenvironments. This advancement has the potential to enhance the predictive drug screening process in personalized cancer therapy significantly.

The bioaugmentation performance is severely reduced in the treatment of high saline pesticide wastewater because the growth and degradation activity of pesticide degraders are significantly inhibited by high salt concentrations. In this study, an artificial halotolerant degrader J9U-MP capable of mineralizing p-nitrophenol (PNP) substituted organophosphorus pesticides (OPs) [e.g., methyl parathion (MP)] was created by integrating a MP-mineralizing pathway into the genome of a salt-tolerant chassis Halomonas cupida J9. MP degradation coupled with stable isotope analysis indicated that J9U-MP was able to metabolize MP as a sole carbon source to finally produce CO 2 and H 2O in high-salt media (up to 120 g/L NaCl). J9U-MP was genetically stable during passage culture and exogenous gene integration did not negatively influence growth and metabolism of J9U-MP. A real-time monitoring system was established with enhanced green fluorescent protein (EGFP) to track the movement and activity of J9U-MP in environmental remediation. A low-oxygen tolerant system was developed by enhancing oxygen utilization, which makes J9U-MP maintain the MP-mineralizing activity under oxygen-limited conditions. More importantly, efficient mineralization of MP by J9U-MP in high saline wastewater was demonstrated. This study highlights that synthetic biology has opened up new avenues for creating stress-resistant pollutants-mineralizing microbes. Competitive advantages of J9U-MP in high-salinity and low-oxygen environments make this degrader suitable for bioaugmentation of pesticide wastewater.

The devastating global toll precipitated by the SARS CoV-2 outbreak and the profound impact of vaccines in stemming that outbreak has established the need for molecular platforms capable of rapidly delivering effective, safe and accessible medical interventions in pandemic preparedness. We describe a simple, efficient and adaptable process to purify SARS CoV-2 virus-like particles (VLPs) that can be readily scaled for manufacturing. A rapid but gentle method of tangential flow filtration using a 100 kDa semi-permeable membrane concentrates and buffer exchanges 0.5 L of SARS CoV-2 VLP containing supernatant into low salt and optimal pH for anion exchange chromatography. VLPs are washed, eluted under high salt, dialysed into physiological buffer, sterile filtered and aliquoted for storage at – 80 0C. Purification is completed in less than two days. A simple quality control process includes Western blot for coincident detection of Spike, Membrane and Envelope protein as a proxy for intact VLP, ELISA to detect conformationally sensitive Spike using readily available anti-Spike and/or anti-RBD antibodies, and negative stain and immunogold electron microscopy to validate particulate, Spike crowned VLPs. This process to purify SARS CoV-2 VLPs for preclinical studies serves as a roadmap for preparation of more distantly related VLPs for pandemic preparedness.

Detergent-mediated viral inactivation is an important process step for ensuring viral safety of parenteral biotherapeutics, including plasma proteins and monoclonal antibodies (mAb). The conventional Triton™ X-100 detergent has ecological toxicity concerns and REACH classification that mandate replacement in the biopharmaceutical industry. Criteria for a replacement detergent include viral inactivation efficacy, acceptable safety and biodegradation profile, process removal and quality suitable for parenteral drug product manufacturing. A non-ionic, C11-15 secondary alcohol ethoxylate, Deviron ® 13-S9 detergent, has been demonstrated to meet the necessary requirements for detergent performance. Benchmarking studies with Triton™ X-100 demonstrate comparable performance with a panel of enveloped viruses in multiple matrices, including human IgG, clarified cell culture harvest, and fractionated plasma. Deviron ® 13-S9 detergent demonstrated viral inactivation efficiency comparable to or better than, Triton™ X-100 detergent, achieving > 5 log reduction values (LRV). Critical micelle concentration (CMC) was determined across different temperatures and media. Deviron ® 13-S9 detergent was demonstrated to be readily biodegradable according to OECD 301B guidelines. Effective removal with typical chromatography processes used in downstream purification was confirmed. These findings support Deviron ® 13-S9 detergent as a viable alternative to Triton™ X-100 detergent, ensuring robust viral inactivation, environmental compatibility, and alignment with regulatory requirements.

Scale-up of microbial bioprocesses to production scale is accompanied by significant changes in growth conditions. One aspect are rapidly changing substrate concentrations due to prolonged mixing times. The impact of substrate gradients on the physiology of Saccharomyces cerevisiae has been studied for glucose but no other industrially relevant carbohydrate. With different transport mechanism, non-glucose sugars will lead to a different intracellular response. Here we studied the response of S. cerevisiae to gradients for four different sugars: glucose, fructose, sucrose and maltose. To study the impact of the carbon sources as well as the large-scale gradients, steady-state and dynamic feast/famine cultivation conditions were applied. The physiology, intracellular metabolome as well as the proteome were compared. Especially, gradients of maltose lead to a significant decrease in biomass yield. Under dynamic glucose, fructose and sucrose conditions S. cerevisiae was able to maintain the biomass yield of steady-state conditions. Although the physiology was very comparable for these sugars, the intracellular metabolome and proteome changed. The concentration of upper glycolytic enzymes decreased for glucose and maltose (up to -60% and -40% respectively), while an increase was observed for sucrose and fructose when exposed to gradients. At the same time, enzymes of lower glycolysis were increased. Interestingly, common stress-related proteins were decreased during dynamic conditions. The observed adaptations to repeating gradients highlight the importance to study physiology and metabolism under dynamic conditions to obtain results that are relevant for the envisioned large-scale process.

Natural Killer (NK) cells, integral to the innate immune system, are notable in cell therapies because for their applicability in allogeneic treatments, distinguishing them from T cells typically employed in conventional cell therapies. However, their limited half-life poses a challenge for therapy. Although attempts to leverage feeder cells are common, safer methods are needed to mitigate the associated risks. In our study, an upregulation in the expression of 4-1BBL in Colo-205 cells under extracellular stresses such as hypoxia and cytochalasin D was observed. This enhanced binding to the 4-1BB receptors on NK cells promotes proliferation in NK cells. Elevated CD56 expression of a marker strongly linked to NK cell proliferation in co-culture further supports this process. Applying extracellular stressors, specifically hypoxia and cytochalasin D, to Colo-205 cells successfully tailored feeder cells, significantly enhancing NK cell proliferation.

Gamma-aminobutyric acid (GABA) is widely applied in the food and pharmaceutical industries, and glutamate decarboxylase (GAD) is the core enzyme for the biosynthesis of GABA. GAD exhibits high activity but poor stability under acidic conditions, and low activity under neutral conditions. In this study, GAD from Priestia flexa ( PfGAD) with high activity was screened out and characterized, and a variant with significantly improved stability was obtained through molecular modification strategies. We constructed an engineered strain that coexpressed PfGAD and a transporter protein for GABA and L-glutamic acid. Furthermore, the consumption of GABA was suppressed by knocking out the pepD gene, confirming a novel downstream metabolic pathway of GABA in Escherichia coli. The continuous accumulation of GABA in the engineered strain led to a yield of 154.7 g·L -1, with a conversation rate of 100%, meanwhile the cell growth was remained normal. The problem of the trade-off between cell growth and GABA accumulation was solved, which was one of the urgent problems according to previous reports. Our study provides important insights into the optimization of GABA production through enzyme engineering and strain modification, which could have significant implications for the industrial application of GABA.

As a non-destructive sensing technique, Raman spectroscopy is often combined with regression models for real-time detection of key components in microbial cultivation processes. However, achieving accurate model predictions often requires a large amount of offline measurement data for training, which is both time-consuming and labor-intensive. In order to overcome the limitations of traditional models that rely on large datasets and complex spectral preprocessing, in addition to the difficulty of training models with limited samples, we have explored a genetic algorithm-based semi-supervised convolutional neural network (GA-SCNN). GA-SCNN integrates unsupervised process spectral labeling, feature extraction, regression prediction, and transfer learning. Using only an extremely small number of offline samples of the target protein, this framework can accurately predict protein concentration, which represents a significant challenge for other models. The effectiveness of the framework has been validated in a system of Escherichia coli expressing recombinant ProA5M protein. By utilizing the labeling technique of this framework, the available dataset for glucose, lactate, ammonium ions, and optical density at 600 nm (OD600) has been expanded from 52 samples to 1302 samples. Furthermore, by introducing a small component of offline detection data for recombinant proteins into the OD600 model through transfer learning, a model for target protein detection has been retrained, providing a new direction for the development of associated models. Comparative analysis with traditional algorithms demonstrates that the GA-SCNN framework exhibits good adaptability when there is no complex spectral preprocessing. Cross-validation results confirm the robustness and high accuracy of the framework, with the predicted values of the model highly consistent with the offline measurement results.

Substantial wealth of knowledge about anaerobic fungi has accumulated in recent years, which may guide the attention of the biotechnology-oriented scientists towards the possible exploitation of their fascinating capabilities. Very efficient, unique and complex enzyme systems of anaerobic fungi play determining role in the conversion of lignocellulosic fodder to milk and meat in mammalian herbivores. Mitigation of the concomitant greenhouse gas emission by ruminants is a major environmental, climate change issue. In turn, controlled management of the inter-kingdom syntrophic co-operations among the eukaryotic anaerobic fungi, bacteria and archaea can lead to the production of valuable bio-fuels, e.g. bio-methane, bio-hydrogen, bio-ethanol, and organic acids, the latter could serve as building blocks in numerous biosynthetic processes in circular bioeconomy.