Chemical Engineering

Permanent URI for this collectionhttps://uwspace.uwaterloo.ca/handle/10012/9904

This is the collection for the University of Waterloo's Department of Chemical Engineering.

Research outputs are organized by type (eg. Master Thesis, Article, Conference Paper).

Waterloo faculty, students, and staff can contact us or visit the UWSpace guide to learn more about depositing their research.

Browse

Now showing 1 - 17 of 17

Development of a Serum-Free Chemically Defined Medium for Adherent and Suspension Culture
(University of Waterloo, 2022-01-07) Logan, Megan; Aucoin, Marc
The development of suspension cell lines is a sought after holy grail for bioprocess development. Suspension cell lines allow for easier scale up and better volumetric productivity for most biotherapeutics, including vaccine manufacturing. Various cell lines have readily adapted to suspension growth through the modification of the cell culture medium, but Vero cells have resisted the trend of becoming a suspension cell line. For Vero cells and other cell lines that remain anchorage-dependent, researchers have found methods such as roller bottles and microcarriers to scale up production of biotherapeutics. Although these technologies have enabled the use of anchorage-dependent cell lines for biomanufacturing, the industry prefers suspension cultures and will design new cell lines, or design their processes around existing suspension cell lines to avoid the extra complexity of using anchorage-dependent cell lines. Medium development has been the main method to adapt cells from adherent culture to suspension. This was first done by reducing the amount of serum that is supplemented in cell culture media. Serum is animal derived and contains many essential nutrients, growth factors, and adhesion proteins that traditional basal mediums do not contain. While serum has been essential for the establishment of many cell lines, it has safety concerns when producing biotherapeutics. Since serums are animal-derived, they can contain viruses, prions, or be contaminated with chemicals that the animal has ingested. Not only do these safety concerns create problems for the biomanufacturing industry, but also the lot-to-lot variability of serums can lead to poor product consistency. To overcome these challenges, the industry has developed animal component-free medium supplements that can contain plant hydrolysates or peptones. These are undefined bioactive fractions of hydrolyzed plant proteins, and while they do not have the same safety concerns as animal derived materials, they also suffer for lot-to-lot variability since the raw materials can depend on soil conditions, weather patterns, and pesticide use. Nevertheless, industry has overcome these challenges by developing chemically defined media. Every component's concentration in this medium is known and therefore this reduces the lot-to-lot variability. Additionally, many of the compounds that are commonly used in chemically defined medium can be manufactured using animal origin-free materials. As cell culture media have eliminated serum, cells have become more anchorage-independent and have been slowly adapted to suspension growth. Although this has not been historically true for Vero cells, which have been grown in tight cell aggregates in serum-free medium. More recently, however, Vero cells have been successfully grown as single cell suspensions using two different undisclosed medium formulations that contain undefined plant hydrolysates or peptones. While this was a major achievement, the medium formulations are unknown, and they contain undefined proteins which can possibly suffer lot-to-lot variability. This thesis seeks to develop a chemically defined medium to support suspension Vero cells and to identify the transcriptomic differences between adherent and suspension Vero cells. This work began by creating a data set of the previously reported medium formulations for Vero, CHO, HEK293, and various other commercial cell lines for adherent and suspension cell culture. Classic basal mediums that require serum supplementation were first compared to serum-free medium formulations to identify compounds that were added to media to replace serum. These compounds included glucose, amino acids, vitamins, trace metals, lipids and fatty acids, along with growth factors and proteins. A second comparison was done within the serum-free medium formulations to compare adherent versus suspension mediums. Suspension media were found to be more enriched that the adherent media and contained higher concentrations of amino acids and fatty acids. This review resulted in theoretical chemically defined, serum-free medium formulations that could support the growth of adherent and suspension commercial cell lines. From this data set, a design space was laid out for the creation of a chemically defined medium formulation that supports Vero cell suspension culture. Through this work, recombinant epidermal growth factor was found to be essential for Vero cell proliferation along with the addition of trace metals, lipids, amino acids, and vitamins. Even without any serum, Vero cells continued to adhere to non-tissue culture treated flasks, while it was found that other cell lines (CHO-K1, HEK293T and MDCK) could grow as single cell suspension in the same novel medium. In an effort to coerce the Vero cells to grow in suspension the concentration of calcium and magnesium was reduced 10x. While this did cause the Vero cells to detach and form a single cell suspension, the growth rate dramatically decreased. Nevertheless, this work demonstrated that a chemically defined medium can be developed for Vero cell suspension, although compounds need to be added to ensure that Vero cells continue to grow. To further investigate the cause of the low growth rate of the suspension Vero cells, RNA-seq was performed to compare the suspension Vero cells to Vero cells grown adherently in the chemically defined medium, and Vero cells grown in DMEM/F12+10\% FBS. This data set demonstrated that the suspension Vero cells had down-regulated cell cycle genes, and had begun to express kidney-associated genes. Since Vero cells were originally isolated from a female Green African monkey's kidney, it is hypothesized that the suspension Vero cells were reverting back to kidney cells. Key genes that were found to be differentially expressed by the suspension Vero cells included \textit{tgfb1}, \textit{c-myc}, and genes that are associated with the epithelial-mesenchymal transition. Finally, different methods for improving the medium formulation were compared in the final chapter. Comparing a literature search, NMR metabolite analysis and RNA-seq transcriptomics found that the transcriptomic data provided the most insight on the compounds that were beneficial for cell growth. The literature did not have enough specific information about Vero cell metabolism and the compounds that were identified using this method did not result in significant improvements in the medium formulation. NMR was more specific for Vero cells, but because it can only track relatively high concentrations (micromolar) of metabolites, changes in vitamins and growth factors could not be tracked. The metabolite analysis did show a disfunctional tricarboxylic acid cycle, and some compounds that were identified using the analysis did significantly improve the cell's growth rate. Overall, the compounds that were identified using the transcriptomic data had the largest effect on the growth rate. Through RNA-seq analysis, we identified retinyl acetate, progesterone, \textbeta{}-estradiol and prostaglandin E\textsubscript{2} as growth enhancing compounds. Using the compounds identified through RNA-seq, NMR and a literature review, the doubling time of Vero cells was reduced from 38 hours to 32.1 hours, which is better than the animal component-free commercially available media currently on the market today.
Establishing the Virucidal Properties of a Copper-Nickel-Zinc Alloy using a Genetically Modified Insect Virus
(University of Waterloo, 2019-08-28) Walji, Sadru-Dean; Aucoin, Marc
Healthcare associated infections (HAIs) cause a significant financial burden on the healthcare industry. To reduce incidence of HAIs, strategies for cleaning and infection control have been developed to target common transmission routes. Surfaces that are frequently touched, so called high-touch surfaces, have been identified as a key transmission route. High-touch surfaces are cleaned inconsistently, and frequent disinfection of these surfaces is impractical. Copper and its alloys have been shown to have contact biocidal properties and pose a solution to inconsistent and infrequent cleaning of high-touch surfaces. However, the virucidal properties of these surfaces are poorly understood due to variability in literature methods. The goals of this thesis are assess the virucidal properties of a copper(65)-nickel(18)-zinc(17) alloy designed to replace stainless steel high-touch surfaces, investigate factors that may diminish the virucidal properties, evaluate virucidal activity of each alloy component, and characterize leaching of metals from the alloy. The copper alloy was shown to have a strong virucidal activity under clean and moderate soiling conditions (>4-log) for virus droplets or dried virus onto the surface. Multiple exposures of the surface to virus found that the surface was unable to inactivate virus droplets (<1-log) while dried virus was repeatably inactivated (>3-log), regardless of no or moderate soiling. Heavy soiling reduced inactivation below an acceptable efficacy threshold (<1-log). Copper, nickel, and zinc were identified as the primary metals being released (leached) by the surface and causing virucidal activity. These metals were further investigated as ion solutions. Virucidal tests of copper, nickel, and zinc ions found that copper and nickel were significantly virucidal (Cu,Ni p<0.05; Zn p>0.1). Concentration of the leached metal ions was dependent on the solution applied to the surface. The driving force behind leaching could not be identified but osmolarity, chlorine content, and protein load of the solution were ruled out.
Examination of binding elements and conditions of Cryptosporidium parvum oocysts to assess its detection potential in water
(University of Waterloo, 2022-09-29) Rodriguez Ruiz-Andino, Irene; Anderson, William; Aucoin, Marc
Cryptosporidium parvum is an intestinal parasite that can be spread through environmental and recreational waters, most often in the form of oocysts. As an oocyst, this parasite is resistant to chlorine disinfection, and it is the known cause of the diarrheal disease Cryptosporidiosis. In this thesis, binding elements like antibodies and aptamers are studied and compared with the purpose of assessing their binding under different conditions and therefore their potential as biorecognition elements to detect C. parvum in water. Several methods have been used to fulfill this purpose. Flow cytometry and imaging flow cytometry was used to compare and assess the binding of commercial fluorescently labelled antibodies, and aptamers reported in literature, under different conditions such as pH, inactivation procedures or heat-inactivation contact time. Horseradish peroxidase (HRP)-labelled antibodies was used to assess the colorimetric potential for Cryptosporidium detection both as a part of an in-filter detecting scheme and to assess the binding differences between live and inactivated Cryptosporidium. It was found that antibodies had better binding affinity than aptamers for Cryptosporidium that was inactivated under different methods such as desiccation, heat-treatment, freeze-treatment, and formalin-treatment. Furthermore, aptamers show an increase of non-specific binding under low pH conditions. An in-filter detection method involving the preincubation of HRP-labelled antibodies with the target and passing this solution through a syringe filter followed by the addition of 3,3',5,5'-Tetramethylbenzidine (TMB), which could allow visual observation of a color change, was also explored. This biosensor was developed as part of a detection scheme to avoid any detection sensitivity loss due to poor recovery from filters and was meant to help make detection simpler for the public. The limit of detection of this approach was found out to be 105 oocysts/mL . While this limit of detection is rather high, improvements could be made by changing the labelling of the antibodies or by changing the type of antibodies.
From Metabolite Concentration to Flux – A Systematic Assessment of Error in Cell Culture Metabolomics
(University of Waterloo, 2016-08-05) Sokolenko, Stanislav; Aucoin, Marc
The growing availability of genomic, transcriptomic, and metabolomic data has opened the door to the synthesis of multiple levels of information in biological research. As a consequence, there has been a push to analyze biological systems in a comprehensive manner through the integration of their interactions into mathematical models, with the process frequently referred to as “systems biology”. Despite the potential for this approach to greatly improve our knowledge of biological systems, the definition of mathematical relationships between different levels of information opens the door to diverse sources of error, requiring precise, unbiased quantification as well as robust validation methods. Failure to account for differences in uncertainty across multiple levels of data analysis may cause errors to drown out any useful outcomes of the synthesis. The application of a systems biology approach has been particularly important in metabolic modeling. There has been a concentrated effort to build models directly from genomic data and to incorporate as much of the metabolome as possible in the analysis. Metabolomic data collection has been expanded through the recent use of hydrogen Nuclear Magnetic Resonance (1H-NMR) spectroscopy for cell culture monitoring. However, the combination of uncertainty from model construction and measurement error from NMR (or other means of metabolomic) analysis complicates data interpretation. This thesis establishes the precision and accuracy of NMR spectroscopy in the context of cell cultivation while developing a methodology for assessing model error in Metabolic Flux Analysis (MFA). The analysis of cell culture media via NMR has been made possible by the development of specialized software for the “deconvolution” of complex spectra, however, the process is semi-qualitative. A human “profiler” is required to manually fit idealized peaks from a compound library to an observed spectra, where the quality of fit is often subject to considerable interpretation. Work presented in this thesis establishes baseline accuracy as approximately 2%-10% of the theoretical mean, with a relative standard deviation of 1.5% to 3%. Higher variabilities were associated primarily with profiling error, while lower variabilities were due in part to tube insertion (and the steps leading up to spectra acquisition). Although a human profiler contributed to overall uncertainty, the net impact did not make the deconvolution process prohibitively imprecise. Analysis was then expanded to consider solutions that are more representative of cell culture supernatant. The combination of metabolites at different concentration levels was efficiently represented by a Plackett-Burman experiment. The orthogonality of this design ensured that every level of metabolite concentration was combined with an equal number of high and low concentrations of all other variable metabolites, providing a worst-case scenario for variance estimation. Analysis of media-like mixtures revealed a median error and standard deviation to be approximately 10%, although estimating low metabolite concentrations resulted in a considerable loss of accuracy and precision in the presence of resonance overlap. Furthermore, an iterative regression process identified a number of cases where an increase in the concentration of one metabolite resulted in increased quantification error of another. More importantly, the analysis established a general methodology for estimating the quantification variability of media-specific metabolite concentrations. Subsequent application of NMR analysis to time-course data from cell cultivation revealed correlated deviations from calculated trends. Similar deviations were observed for multiple (chemically) unrelated metabolites, amounting to approximately 1%-10% of the metabolite’s concentration. The nature of these deviations suggested the cause to be inaccuracies in internal standard addition or quantification, resulting in a skew of all quantified metabolite concentrations within a sample by the same relative amount. Error magnitude was estimated by calculating the median relative deviation from a smoothing fit for all compounds at a give timepoint. A metabolite time-course simulation was developed to determine the frequency and magnitude of such deviations arising from typical measurement error (without added bias from incorrect internal standard addition). Multiple smoothing functions were tested on simulated time-courses and cubic spline regression was found to minimize the median relative deviation from measurement noise to approximately 2.5%. Based on these results, an iterative smoothing correction method was implemented to identify and correct median deviations greater than 2.5%, with both simulation and correction code released as the “metcourse” package for the R programming language. Finally, a t-test validation method was developed to assess the impact of measurement and model error on MFA, with a Chinese hamster ovary (CHO) cell model chosen as a case study. The standard MFA formulation was recast as a generalized least squares (GLS) problem, with calculated fluxes subject to a t-significance test. NMR data was collected for a CHO cell bioreactor run, with another set of data simulated directly from the model and perturbed by observed measurement error. The frequency of rejected fluxes in the simulated data (free of model error) was attributed to measurement uncertainty alone. The rejection of fluxes calculated from observed data as non-significant that were not rejected in the simulated data was attributed to a lack of model fit i.e. model error. Applying this method to the observed data revealed a considerable level of error that was not identified by traditional χ2 validation. Further simulation was carried out to assess the impact of measurement error and model structure, both of which were found to have a dramatic impact on statistical significance and calculation error that has yet to be addressed in the context of MFA.
Generation and Characterization of Dual-Fluorescent Influenza Virus-like Particles (VLPs) in Insect Cells
(University of Waterloo, 2019-01-04) Ramirez, Eduardo; Aucoin, Marc
The Baculovirus Expression Vector System (BEVS) has been widely used to produce recombinant proteins, especially for complex biopharmaceuticals. In recent years, the insect-baculovirus system has proved its value as a robust production platform. In this regard, multiple vaccines have been commercialized such as FluBlok® using this system. The work presented here aims to further explore their use on the production of influenza virus-like particles (VLPs). The goal is to successfully produce fluorescent influenza particles that behave similarly to native particles and that allow the purification of viral particles. Four different genetic constructs were investigated for the expression of dual-fluorescent influenza VLP. Individual influenza genes were fused to different fluorescent proteins. Hemagglutinin was fused with enhanced green fluorescent protein (GFP), whereas, M1 was fused to a red fluorescent protein (RFP). The use of monomeric or multimeric RFP was studied, as well as, the fusion site to M1. HA-GFP was placed under the baculovirus p10 promoter, while M1-RFP was under the control of the baculovirus polh promoter. Sf9 cells were used to generate baculovirus vectors that were later used to infect Sf9 cells and produce the influenza VLPs. Flow cytometry and confocal microcopy were used to detect individual levels of influenza proteins and their localization. within the cell. A proposed downstream process consisting of a 2-step concentration procedure and a density gradient purification for influenza VLPs was evaluated by tracking HA activity and baculovirus presence. Purification of VLPs and baculovirus was achieved by using iodixanol and sucrose density gradients, an undesirable result since baculovirus is the main contaminant produced in the production process. Further sample analysis of purified fluorescent particles was performed by flow cytometry. Our findings showed that recombinant baculovirus driving the expression of M1 fusions, with either monomeric or tetrameric RFP, in the 3’ site presented red fluorescence localized at the center of the cell. The same behavior was observed when a monomeric RFP was fused at the 5’ site. Further purification of fluorescent VLP constructs showed lower HA activity of fluorescent constructs when compared with a non-fluorescent VLPs. Moreover, iv similar migration patterns were obtained via sucrose density gradient. However, the need to add more purification procedures such as ion exchange chromatography are needed to fully isolate influenza VLP particles. Lastly, a flow cytometry protocol to study purified particles was evaluated. Nano-sized fluorescent particles were detected in purified samples, which increased particle counts in samples with higher concentrations of HA activity and baculovirus. However, our analysis detected mainly coincidental events. Single particle events overlapped with signal noise, therefore an increase in the inherent fluorescent of viral particle should be explored to differentiate single particles from signal noise.
An In-Depth Look at Repeated Propagation of Multiple Viruses Simultaneously in Insect Cell Culture
(University of Waterloo, 2020-01-22) Chakraborty, Madhuja; Aucoin, Marc
It is well known that the insect cell-baculovirus expression vector system (IC-BEVS) can be used for the production of various proteins of interest and virus-like particles (VLPs). In the past, many higher order proteins and protein complexes like antibodies, viral vectors and VLPs have been produced using this system. For the production of recombinant proteins of interest or VLPs, one or more baculovirus constructs, each carrying one (monocistronic) or more (polycistronic) gene(s) of interest can be introduced to insect cells. Three common expression systems for protein production mainly include infecting insect cells with a number of baculovirus constructs each carrying a gene of interest (coinfection), infecting insect cells with a single baculovirus construct carrying multiple genes of interest (coexpression), or a combination of the two systems. Although coexpression is the ultimate goal, there are many reasons coinfection is still used. Coinfection allows a degree of flexibility that is not as easily achieved through coexpression. The hypothesis behind this work is that, the virus ratios or the population distribution remains constant upon repeated propagation. Here, we are seeking to avoid coinfection at larger scale by amplifying viruses together at smaller scale. By infecting cells with baculovirus coding for reporter proteins, eGFP and mKate2, at high and low MOIs we can track the virus population being amplified and be confident that the ratios are preserved or will require adjustment upon repeated amplification. We can then use this knowledge to improve the production of influenza A VLPs, a more industrially relevant system. Different assays and/or techniques were used to assess the virus population distribution over repeated propagation at high or low MOI, in addition to purification and analysis of influenza VLPs. It was concluded that repeated amplification at high MOI changed the population distribution, and also resulted in significant drops in the infectious virus titer. Amplifications at low MOI improved the virus titer, and not much variation was observed in the distribution over subsequent passaging. That being said, it was found that ultracentrifugation alone did not have the resolution to purify VLPs, and additional purification steps must be employed to produce VLPs for biopharmaceutical applications.
Investigating the Influence of Bacterial Cell Characteristics on M13 Phage Infection Process
(University of Waterloo, 2024-08-29) Haghayegh Khorasani, Seyedeh Sara; Aucoin, Marc; Ingalls, Brian
Microbial communities are fundamental to ecosystem health and biodiversity, affecting environments from soil to human microbiomes. Bacteriophages, or phages, are vital components of these communities, shaping bacterial dynamics and genetic diversity through mechanisms like gene transfer. Traditional population-level studies, while informative, can obscure the detailed behaviors and interactions present at the individual cell level. This research seeks to mitigate this oversight by applying single-cell analysis techniques to explore the M13 phage infection process. Focusing specifically on the interactions between the M13 bacteriophage and E. coli, this research employs time-lapse microscopy to investigate how individual bacterial cell characteristics— size, elongation rate, and spatial positioning—impact phage infection susceptibility. The experimental approach incorporates both microfluidic devices and agar pads to compare the effects of direct phage introduction versus in-situ phage production within mixed bacterial cultures. Image processing was conducted using the Ilastik and CellProfiler software, extracting vital cellular metrics, such as size, shape, elongation rate, and spatial distribution, for analysis. Subsequent post-processing, performed with custom MATLAB scripts, generated lineage trees for individual cells, enabling tracking and analysis of cellular behavior over time. Experimental results demonstrate that E. coli cells exhibiting higher elongation rates and larger sizes were notably more susceptible to M13 phagemid infection. This correlation underscores the significance of physical and physiological cell properties in the infection process. Moreover, this research extends its analysis through computational simulations employing the CellModeller platform, to investigate the M13 bacteriophage infection process beyond what is observable in laboratory experiments. The simulations are particularly concentrated on assessing how variations in phage diffusion rates impact the spatial patterns of infection, especially regarding the proximity of infected cells to those producing phages. The simulation results from this study highlight that an increase in the phage diffusion rate leads to a decrease in the distance between infected cells and those producing phages, suggesting that higher diffusion rates facilitate wider spread and more uniform distribution of the phage within the bacterial population. This pattern is consistent with the hypothesis that phage mobility plays a critical role in the dynamics of infection spread.
Investigation of excipients for the stabilization of HSV-2 vaccine candidate ACAM529
(University of Waterloo, 2018-05-01) Manalil, Julia; Aucoin, Marc
Herpes simplex virus type 2 (HSV-2) is a highly infectious pathogen that causes genital ulcerative disease and it affects millions of people worldwide. ACAM529 is a promising new live replication-deficient vaccine candidate against the virus. Vaccine stability during the upstream production processes is an issue so there is a great need for developing an effective stabilization buffer that will protect the vaccine during the process. The purpose of this work was to study the characteristics of ACAM529 and identify excipients that would protect the vaccine from inactivation in liquid solution short-term. To establish a baseline of ACAM529 stability for further improvements, three lots of purified ACAM529 with differing purification protocols were characterized: Lot A, Lot B and Lot C. Plaque assays were used to determine the titer of each lot, SDS-PAGE was used to analyze the albumin content, and transmission electron microscopy (TEM) was used to image the virus. Lot C was found to have the highest titer at 5.2 ± 0.6 x 10e7 PFU/ml and albumin concentration at approximately 8 g/L compared to Lot A (3.4 ± 0.6 x 10e6 PFU/ml and 0.5-1 g/L, respectively) and Lot B (2.1 ± 0.4 x 10e7 PFU/ml and <0.13 g/L, respectively). TEM imaging revealed a high prevalence of unenveloped compared to enveloped virus and some clumping of virus particles, however no differences could be identified between the lots. In addition, a time course assay was developed to study the stability of ACAM529 over 120h. Lot C was found to be more stable than Lot A and Lot B, especially at lower temperatures. Lot A and Lot B showed similar levels of stability at 2-4°C and 25°C. To screen for ACAM529 stabilizers in liquid solution, various defined and complex solutions used in cell culture for virus production were studied using a time course assay. The addition of 0.01%-10% FBS to the reference buffer resulted in improvements to Lot B stability at 27°C, with minimal titer loss after 120h occurring with 1% FBS. The addition of 0.05-5 g/L rHSA to the reference buffer for Lot B improved viral stability at 2-4°C in a concentration-independent manner. However, at 27°C, 0.5-5 g/L rHSA significantly reduced the stability, where increasing concentrations of rHSA lead to decreasingly stable virus. Low concentrations of rHSA (0.05 g/L) improved stability at 27°C. The addition of 10% conditioned OptiPro SFM resulted in a slight improvement to stability over a period of 120 hours at 27°C, however the same amount of fresh medium resulted in no change to stability. The addition of 0.1x-1x CLC supplements to the buffer resulted in dramatic decreases to ACAM529 stability at 27°C, especially in the first 24h. Lastly, Pluronic F68 (PF68) at concentrations below the critical micelle concentration (CMC), significantly improved viral titers after 120h, with maximal stability achieved at a concentration of 0.1% PF68. Lastly, other excipients that have been implicated in enhancing the stability of protein and lipid-based products were investigated for ACAM529 stabilization. An acidic buffer (pH 5) condition caused an improved initial infectivity of the virus when titered with the plaque assay, however it had a negative impact on long-term stability compared to buffers in the range of pH 6-7. The replacement of 10% (w/v) sucrose with 10-30% (w/v) trehalose in the reference buffer did not result in changes to viral stability at pH 6-7 at 27°C. Also, the addition of 0.01-0.1 M L-arginine (along with the resulting increase in buffer ionic strength with the addition of Cl- ions) and 0.1x-1x non-essential amino acid solution (NEAA) did not significantly affect the stability of ACAM529. This work showed that it was possible to improve the stability of ACAM529 in the current liquid buffer. The highest level of stability was achieved with the addition of 1% FBS. The addition of 0.05 g/L rHSA or 0.1% PF68 also significantly improved the virus stability and these compounds have the added benefits of having a defined chemical composition compared to FBS and being a safer alternative.
Partial Characterization of Yeast Hydrolysates for Insights on Chemically Defined Media for Sf-9 Insect Cells
(University of Waterloo, 2017-10-18) Quattrociocchi, Marco; Aucoin, Marc
Yeast extract, or other varieties of yeast hydrolysate, have been used in combination with a lipid supplement to completely eliminate the need for fetal bovine serum during in vitro cultivations of insect cells. This has made the process cheaper and much more attractive for large industrial uses due to the higher availability of yeast extract over serum. While yeast extract represents a vital step forward in culture media, the gold standard of a chemically defined media with the same growth parameters has not been achieved. To better understand the hydrolysate composition, it was subjected to trace mineral analysis via ICP-OES while H-NMR was used to quantify the detectable organic compounds before, during, and after acid hydrolysis. Trace mineral analysis showed that yeast extract was the sole provider of silicon, chromium, nickel and vanadium, as well as a majority provider of zinc, manganese, copper, cobalt and iron. While silicon has no known function in metabolism, the other elements are of considerable interest complicated by their contributions to detrimental free radical reactions. Yeast extract also provides antioxidants to help combat some of these reactions while possibly assisting in the trace metal dissolution due to the presence of several organic molecules which function as chelators. Through the hydrolysis studies, the yeast extract was shown to be composed of peptides heavily consisting of glycine, lysine and proline. Yeast extract also contains a significant amount of nucleic material, mostly from RNA due to the presence of uridine. Further, several B-vitamins were quantified, including significant amounts of B3 and myo-inositol, each more than 10x their current respective defined supplementation. Metabolic studies of Spodoptera frugiperda Sf-9 cell cultures in media containing hydrolysates helped to illuminate which compounds are active metabolites over the initial growth phase as well as determining any compounds which become completely exhausted. However, consistent cell growth after the commercial media was diluted sufficiently over 3 passages was only obtained with one of the yeast extract lots studied, possibly highlighting deeper concerns with the manufacturing process overall. Along with a thorough literature review, all of this data was combined into a chemically defined supplement to replace the yeast extract. No significant growth was obtained once the commercial media had been diluted from the system over several passages. This suggests that only a small amount of the crucial growth promoting compound is required, but more rigorous investigations (e.g. GC-MS, fractionation analysis) are still required to pinpoint its identity.
Supplementation Strategies for Tuning Glycosylation of Monoclonal Antibodies and Enhancing Growth in Mammalian Cell Culture by Omics Analysis
(University of Waterloo, 2018-04-17) Blondeel, Eric; Aucoin, Marc
Two fundamental objectives of bioprocess engineering are to increase productivity and improve product quality. Glycosylation is a critical and variable factor in the quality of several therapeutic proteins, particularly those with immune-system interactions such as monoclonal antibodies (mAbs). In this thesis, supplementation of nutrients to growth medium of CHO1A7 mammalian cell cultures are examined towards enhancing cell growth, by increasing peak cell density, and improving product quality, by tuning glycosylation of EG2-hFc heavy-chain camelid antibodies. Targeted profiling via 1D-1H-NMR metabolomics and differential expression analysis via 2D-DIGE proteomics, elucidate factors to create a nutrient cocktail to enhance culture growth. Eight target nutrients corresponding with five identified metabolic systems for CHO cells including anaplerotic TCA-replenishment; NADH/NADPH replenishment; tetrahydrofolate cycle C1 cofactor conversions; limitations to lipid synthesis; and redox modulation; resulted in a ~75% improvement to peak cell densities. Towards improving product quality, nucleotide-sugar precursors, capable of shifting glycan distributions were supplemented to growth medium to tune glycosylation of EG2-hFc towards a single G0 glycoform. Growth inhibition from glucosamine-based precursors was mitigated given a priori knowledge from metabolomic analysis of the system – identifying cytosolic acetyl-CoA as a sensitive metabolic pool for CHO1A7 cell growth. Additional nucleotide-sugar precursor nutrients were examined to better resolve conflicting reports of effects to glycan distributions. These conflicts are subsequently attributed to five key factors: differences across cell platforms; differences between glycan sites of expressed proteins; the fermentation and sampling timeline; glutamine levels; finally, no standardized metrics for reporting shifts in glycan distributions with respect to controls.
Tools for Manipulation of Microbial Communities through Bacterial Conjugation
(University of Waterloo, 2025-01-06) Yip, Aaron; Ingalls, Brian; Aucoin, Marc
Microbial communities are ubiquitous across the planet and play essential roles in numerous aspects of society including human health, agriculture, food production, and biodegradation. Controlled manipulation of natural microbial communities by introduction of biological agents is a promising option for precise, in situ microbiome editing. Bacterial conjugation is a well-studied mechanism for gene exchange (horizontal gene transfer) between bacteria in physical proximity to one another. Conjugation can be repurposed as a gene delivery technique into microbial communities and holds potential for addressing problems of environmental and clinical relevance, including the biodegradation of pollutants such as microplastics and delivery of precision antimicrobials. This thesis begins with a proof-of-concept study in which recombinant DNA is delivered to bacteria in wastewater via conjugation to enable the degradation of polyethylene terephthalate (PET), a major component of global plastic pollution. Using a broad-host-range conjugative plasmid, we enabled expression of FAST-PETase in various bacterial species from municipal wastewater, achieving substantial degradation of both commercial PET film and post-consumer PET products under laboratory conditions. Next, the thesis builds computational tools and methodologies to support model-based design of in situ gene delivery into microbial communities through conjugation. Prior to starting model development, a review of practices for calibrating spatio-temporal models to microscopy data was conducted, which demonstrated the need for a more formal process for generating predictive models. Drawing on practices from ecology, new strategies for systematic model validation were proposed. Some of these strategies were then implemented into a model calibration pipeline based on Pattern-Oriented-Modelling and Bayesian parameter inference, and the pipeline was demonstrated by fitting biophysical parameters in an agent-based model to time-lapse microscopy data. The calibrated model was able to reproduce several patterns in microcolony formation that were observed experimentally, but did not fully replicate patterns associated with colony shape. Finally, a single-cell-based approach to characterizing conjugation in microfluidic environments was developed to investigate spread of the previously developed conjugative plasmid that enabled expression of FAST-PETase. Although the plasmid could spread through conjugation, cells bearing the conjugative plasmid tended to get outcompeted for space by the faster-growinghealthier recipient population. The tendency of cells to self-orient in the direction towards the exit of the microfluidic traps through biomechanical processes also reduced conjugation efficiency. The previously developed model calibration pipeline was then applied to calibrate an agent-based model for conjugation to the microscopy data collected. Afterwards, the calibrated model was used to characterize how initial conditions and spatial factors influenced spread of conjugative plasmids in enclosed microenvironments. Collectively, these studies enhance understanding of engineering microbial communities through conjugation, offering novel solutions for plastic waste degradation and advancing model-based design of gene delivery into microbial communities.
Toward Optimization of the Baculovirus Expression Vector System - Development of Genetic Tools to Improve Biologics Production
(University of Waterloo, 2021-08-25) Bruder, Mark; Aucoin, Marc
Small-molecule drugs have dominated the pharmaceutical industry and physician's prescription pads since the beginning of modern medicine. The most recent decades, however, are most aptly characterized as the era of biologics. Indeed, 7 of the top 10 revenue generating therapeutics in 2020 were biologics, and although they are yet to be counted among the most commonly prescribed drugs, the drum beat of new product approvals and demand for biologic therapeutics is intensifying. Despite the potential of biologics as therapeutics, one of the factors that currently hinder their use is exorbitant cost of production. In preceding years, important developments in the optimization of media, feeding strategies, and downstream processing have led to signifi cant improvement in the yield and decreased production cost of recombinant therapeutics. The most recent decade, however, has witnessed a revolution in the field akin to the systems biology approach toward rational design practiced for improving microbial production hosts; enabled by advancements in whole-genome sequencing, genome-scale models, and development of sophisticated genetic engineering tools, a new wave of engineered production hosts and platforms with improved or completely novel biochemical properties are being developed, leading to improved product yield and quality, and decreased production costs. The baculovirus expression vector system (BEVS) is an established platform for the manufacture of recombinant proteins, viral vaccines, and gene therapy vectors. Despite the fi rst recombinant protein being produced in the BEVS in the early 1980s, much of the intervening years has seen it utilized predominantly as a research tool in academic laboratories rather than as a commercial manufacturing platform. Consequently, relatively little attention has been devoted to its improvement as a production platform. Nevertheless, several BEVS-manufactured vaccines and therapeutics have recently been licensed for use in animals and humans, signifying that it may yet nd mainstream use as a commercial manufacturing platform. Although the BEVS boasts many features that make it attractive as a manufacturing platform, to realize its full potential, intrinsic limitations must be addressed: the lytic infection cycle and resulting short bioprocess duration can limit overall yield of recombinant proteins, and large amounts of progeny virus, cellular proteins, and debris from lysed cells are contaminants that necessitate extensive purifi cation steps to achieve product purity. Additionally, genome instability remains a major barrier to scalability due to rapid loss of foreign gene expression. Although periodic reports in the literature describe strategies aimed at reducing contaminant progeny virions or improving yield and/or quality of the recombinant protein product by targeted deletion or addition of endogenous and heterologous genes in the baculovirus genome, genomes available commercially remain virtually unmodifi ed. This thesis seeks to address these issues through the development of genetic tools aimed at optimization of the baculovirus genome. We initiated this work by developing a platform for efficient scrutiny of baculovirus genes through targeted gene disruption and transcriptional repression using CRISPR-Cas9 technology. Using cell lines that were developed for constitutive expression of the Cas9 or dCas9 proteins, sequence-specifi c disruption or downregulation of target genes was achieved with efficiencies of up to 99%. The key factors affecting efficiency were choice of promoter for sgRNA expression and spacer sequence selection for gene targeting. CRISPR-mediated gene disruption was more effective than transcriptional repression in all cases. As a result of these fi ndings, we confi rmed sequence-specifi c disruption of the AcMNPV GP64 and VP80 structural proteins for recombinant protein production with reduced baculovirus contamination. Targeting these genes resulted in greater than 94% reduction in budded virus release. Importantly, production of the model cytoplasmic protein GFP and a model virus-like particle based on the HIV-1 Gag protein was not signi ficantly affected by gene disruption, indicating that our approach could be more efficient than previously reported strategies. Next, a microplate-based assay was developed to allow for efficient scrutiny of several baculovirus genes in parallel. The assay involved transfection of Sf9 cells constitutively expressing the Cas9 protein with a plasmid encoding a sgRNA, followed by infection with a recombinant BEV. Expression of a gfp reporter gene and analysis of infectious virus titer in cell culture supernatants were used as analogs for late gene expression and progeny budded virus release, thus providing insight toward the effect of various gene disruptions on the virus infection cycle. The critical factors established in the development of the assay included viable cell density, choice of transfection reagent, the amount of plasmid DNA transfected, the ratio of transfection reagent:plasmid DNA, the time interval between transfection and infection, virus multiplicity of infection, and the time interval between infection and harvest/analysis. The assay was used to scrutinize the effect of disrupting 13 AcMNPV genes, and the results agreed with those previously reported in all cases. Importantly, results could be realized in less than 2 weeks, which represented an improvement in efficiency of up to several months. Finally, bioinformatics was used to select and evaluate baculovirus promoters with different expression characteristics than those routinely employed for foreign gene expression. We assesed the selected promoters by expressing model cytoplasmic and secreted proteins, and provided experimental evidence of the importance of promoter selection for foreign gene expression. We also examined sequence determinants that may be important for late gene transcription and translation initiation on a genome-wide scale.
Towards a Chemically Defined Medium for Sf-9 Cell Culture: Micronutrients Reduce Dependence on Yeast Extract
(University of Waterloo, 2020-01-08) Boegel, Scott; Aucoin, Marc
Spodoptera frugiperda clonal isolate 9 (Sf-9) insect cells in conjunction with recombinant baculovirus are an industrially relevant system for producing biologics. Sf-9 cells are capable of robust high-density growth in single cell suspension. However, unlike many other continuous cell lines, Sf-9 cell culture media remains undefined. Typically, the growth medium requires undefined hydrolysate supplementation (most often yeast extract) in order to support cell proliferation. The lack of chemical definition makes medium and process optimization difficult, leads to batch-to-batch variability, and potentially affects downstream processing. This work aims to combine available information on the composition of yeast extract and the composition of media for other cell lines to reduce the concentration of undefined components (yeast extract) in the medium and elucidate the effects of micronutrient compounds. Utilizing an in-house medium based on the classic IPL-41 medium with yeast extract as the only undefined component, several steps were taken towards chemical definition. Through fortifying the trace metal and vitamin content in the medium and the addition of 11 micronutrients, the yeast extract content was successfully reduced 10-fold (from 4 g/L to 0.4 g/L). Without medium fortification and micronutrient addition, the cells were incapable of growth at low yeast extract concentration. Sf-9 cells adapted to this new medium were capable of long-term consistent growth. Micronutrients of key importance in this medium were identified as glycine betaine, ascorbic acid, and the polyamine putrescine. The presence of glycine betaine (1 mM), ascorbic acid (10 uM), and putrescine (10 uM) improved maximum cell density by 32%, 41%, and 28% respectively in the low yeast extract medium. The role of these micronutrients could be properly investigated only after medium enhancement and yeast extract reduction. Further, this medium was found to be cost-effective compared to commercially available alternatives and the potential for added cost-savings related to lipid supplementation was identified. This enhanced low yeast extract medium could allow for micronutrient and other component investigation with less convolution and is particularly applicable to designed compound screening experiments (e.g. Plackett-Burman). Identification and supplementation of additional required components provided solely by the yeast extract could lead to a chemically defined medium.
Towards the Development of a ϕC31Int-Competent Sf9 Cell Line
(University of Waterloo, 2018-12-05) Brogee, Paul; Aucoin, Marc
The Sf9 cell line, as a component of the Baculovirus Expression Vector System (BEVS), is a popular platform for industrial-scale production of protein products. However, the natural course of the BEVS results in host-cell lysis before many down-stream protein modifications can be made, making it suboptimal for production of complex products such as membrane or glycosylated proteins. Stably transforming a Sf9 cell line for constitutive expression of a gene product can effectively circumvent this issue, however the process of developing these cell lines remains inefficient and time-consuming. Synthetic biology is a field of research that approaches biological problems through the lens of engineering design. With this in mind, a design for a Sf9 line was developed that leverages ϕC31 Integrase technology to facilitate Recombinase Mediated Cassette Exchange (RMCE) in vivo in the Sf9 host cell. RMCE is a type of site-specific DNA recombination that results in the exchange of a DNA cassette from, in this case, a donor plasmid to a genomic locus within the host cell. A pair of fluorescent reporter proteins, mAzamiGreen and mKOk, were tested to determine their suitability for use in confirming successful cassette exchange in Sf9. These were found to be easily differentiable in both mixed culture and co-transfection scenarios. Additionally, some insight was gleaned with respect to media formulation when performing a limiting dilution protocol on Sf9 cells. Culturing cells in a 50/50 Fresh/Conditioned media formulation showed increased cell viability when compared with 100% conditioned media. The addition of FBS drastically improved cell viability after 1 week. However, the difference between 2% and 10% FBS supplementation was not significant. The genetic components of the design were assembled via Splice Overlap Extension PCR and Gibson Assembly protocols. However, complications during the construction of the genetic architecture meant that the design was not seen to completion. Nevertheless, the lessons learned throughout this work will aid in future attempts to develop the proposed system.
Tracking Uptake, Assembly and Egress of ACAM529 Herpes Simplex Virus Type 2 in Complementary Vero Cells for Enhanced Virus Recovery
(University of Waterloo, 2016-01-06) Malenkov, Valentin; Aucoin, Marc
Herpes simplex virus (HSV) type 2 is the cause of one of the most wide-spread human infections in the world, affecting millions of people and circumventing all past attempts to cure or prevent infection. One of the biopharmaceutical companies involved in HSV-2 vaccine development, Sanofi Pasteur, has developed a candidate vaccine based on a live, replication-deficient HSV-2 virus. The “ACAM529” strain is subject to two gene deletions and requires a complementary African Green Monkey Kidney (Vero) cell line (AV529-19) to replicate. A bottleneck exists in the production process which is hypothesized to be the result of virus particles being directed towards both cell-cell and cell-substrate (adherent surface) junctions. The overarching goal of this work was to determine if some mature virus particles localize and egress at substrate surface junctions, becoming inaccessible to traditional extraction methods. To test the hypothesis, an alternate titration method was first developed using induced fluorescence end-point dilution assay with a spectrofluorometer. The second stage of the research involved the use of immunofluorescent staining to locate viral proteins within infected cells in order to inform future process optimization. Mature virus particles were of particular interest and the co-localization of two essential proteins was used to determine their location. A pseudo co-culturing protocol was then developed using cell culture inserts. This system was used in an attempt to determine the extent to which viral budding occurred towards the substrate. The fluorescence which the HSV2012-121 Vero cells displayed on infection with ACAM529 did not allow for effective differentiation between infected and uninfected cells. Treatment of the cells with Sodium Butyrate (2mM NaBu added at the time of seeding) resulted in a 16.93-fold difference between infected and uninfected cells’ fluorescence. Immunofluorescent staining of two virus proteins to determine their co-localization was not successful. Evidence pointed to bleeding of the signals between the 488 and 405nm channels, as well as to a lack of capsid staining in the presence of a tegument of envelope layer. Nevertheless, individual glycoprotein stains were found to be effective and provided good insights into the virus’ behaviour. Lastly, the AV529-19 Vero cells were shown to adhere effectively to insert membranes and form confluent monolayers. Staining and counting of the resulting plaques was used to establish a required initial infection titer between 8.3x104 and 1.2x106 PFU/mL for future cell insert experiments.
Understanding the Sf9 – Baculovirus Expression Vector System: Exploring the Metabolic Basis of the Cell Density Effect through Nuclear Magnetic Resonance Spectroscopy and Flow Cytometry Analysis of Fluorescent Protein Accumulation
(University of Waterloo, 2020-05-15) Pritchard-Oh, Alexander; Aucoin, Marc; Ward, Valerie C.A.
It is well established that the baculovirus expression vector system (BEVS) can serve as an important, effective production system for various biologicals. Knowing that the cells can serve as a production platform is useful, but there are still gaps in the knowledge of how best to manipulate this platform. There is room for improvement in both the upstream (in the design of the viral vector) and the downstream (in the optimization of the operating conditions of the growth culture) portions of the process. This thesis aimed to look at the latter group. In particular, this thesis was aimed to assess some gaps in the knowledge of the metabolic requirements of the cells as they grow and are infected. It was found that higher growth was easily achievable, but growth rate drops beyond a certain cell density regardless of frequency of media exchange. Cell density could be extended to 44 x 106 cells/mL, up from the 12 x 106 cells/mL achieved in batch culture, but that cell growth rate would drop beyond 15 x 106 cells/mL, regardless of the frequency of media replacement. Exploring this idea, it was determined that oxygenation did not appear to be the limiting condition here, as growth rate did not drop in a separate experiment wherein the cells were intentionally deprived of oxygen, down to less than a third of the concentration in the control culture. By using Proton Nuclear Magnetic Resonance (1H-NMR) spectroscopy to measure metabolite levels, eight compounds not studied in the literature were identified to drastically vary over the culture. Fructose, fumarate, hypoxanthine, and succinate all accumulated in the culture while choline, inosine, uracil, and uridine were depleted. While these compounds also did not seem to perfectly correlate with the drop in growth rate, it was hypothesized that they may be important for supporting growth and productivity. Fluorescent proteins were used to give a simple analogue for the productivity of the cells. However, a methodology for quantifying the productivity was needed. As an initial test, 1H-NMR revealed that the composition of the baculovirus stocks were very similar to normal spent media, and with the volumes of baculovirus stock required for addition to the culture will have minimal impact on bulk culture composition (although this does not rule out the influence of minor compounds undetectable via 1H-NMR). However, only the GFP virus was deemed to be useable based on the characteristics of its fluorescence. It was found that as a standard infection progressed, two distinct populations appeared. Testing found that these populations corresponded to healthy and compromised cells, respectively. The average fluorescence of only the healthy population increased with time whereas the average fluorescence of the compromised population was centered on a comparatively low value. Averaged together, this was a net 20% decrease in average fluorescence between 48 and 72 hpi. Individually, however, the healthy cells had roughly 60% higher fluorescence at 72 hpi than at 48 hpi. Therefore, a proposed “improved” gating method was chosen, where only the healthy population was looked at as a reporter for the progression of the infection. Overall, it was deemed that this would provide more useful information about the quality of the infection. Lastly, the goal was to improve the productivity of cell cultures under practical conditions. To this end, baselines for the production of a fluorescent protein were established, and then the methodology developed was used to assess the effect that the compounds of interest identified would have on the productivity of cells infected under suboptimal conditions. It was found that, as expected, the average fluorescence of a culture decreased when moving from low cell-density infection to high cell-density infection conditions (a 50% decrease). Interestingly, at low cell-density, media replacement was detrimental to fluorescent protein production (a 20% decrease) while at high cell-density, media replacement was beneficial (a 33% increase). An experiment was set up according to Placket-Burman design in order to effectively probe the effect of the 8 compounds of interest, and it revealed succinate to have an overwhelmingly negative impact on fluorescent protein production. However, it was suggested that the majority of this effect seemed to have come from pH change rather than inherently from succinate itself.
Use and Control of Co-Expression in the Baculovirus-Insect Cell System for the Production of Multiple Proteins and Complex Biologics
(University of Waterloo, 2016-08-19) George, Steve; Aucoin, Marc
The ease of use and versatility of the Baculovirus Expression Vector System (BEVS) has made it one of the most widely used platforms for recombinant protein production. In the last ten years, the system has even gained commercial acceptance for the production human biologics such as for the production of human papillomavirus vaccine (Cervarix®) and influenza vaccine (FluBlok®). The work presented in this thesis aims to further the utility of this system in coexpressing multiple proteins within a single cell, with the final goal of setting up designed and tightly controlled gene expression schemes within insect cells. In this work we explore the effect of using different Autographa californica multicapsid nucleopolyhedrovirus (AcMNPV) promoters to control the timing and expression of multiple proteins within insect cells. We do this first in a simple two fluorescent protein system where protein and RNA expression levels of these two proteins are tracked over time and complex interaction effects are evaluated. Some of these promoter combinations are then used in a more complex and industrially relevant influenza VLP production system, in which effects such as significant post translation modification, protein trafficking to the cell membrane, insertion of protein into the membrane, and VLP budding from the surface are present. These studies coexpress two influenza proteins – the Hemagglutinin (HA) and Matrix 1 (M1) proteins genetically fused to eGFP and DsRed2 fluorescent proteins to easily track influenza protein localization and expression levels within insect cells, and to track influenza virus-like particles in culture supernatant. Lastly, we examine the effect of coexpressing superoxide dismutase as a helper protein to extend the lifespan of insect cells post-baculovirus infection.

Browse

Browsing Chemical Engineering by Author "Aucoin, Marc"