The Choice of Suitable Online Analytical Techniques and Data Processing for Monitoring of Bioprocesses

Cole-Cole modeling of real-time capacitance data for estimation of cell physiological properties in recombinant Escherichia coli cultivation

Real-time estimation of physiological properties of the cell during recombinant protein production would ensure enhanced process monitoring. In this study, we explored the application of dielectric spectroscopy to track the fed-batch phase of recombinant Escherichia coli cultivation for estimating the physiological properties, viz. cell diameter and viable cell concentration (VCC). The scanning capacitance data from the dielectric spectroscopy were pre-processed using moving average (MA). Later, it was modelled through a nonlinear theoretical Cole-Cole model and further solved using a global evolutionary genetic algorithm (GA). The parameters obtained from the GA were further applied for the estimation of the aforementioned physiological properties. The offline cell diameter and cell viability data were obtained from particle size analyzer and flow cytometry measurements to validate the Cole-Cole model. The offline VCC was calculated from the cell viability % from flow cytometry data and dry cell weight concentration (DCW). The Cole-Cole model predicted the cell diameter and VCC with an error of 1.03% and 7.72%, respectively. The proposed approach can enable the operator to take real-time process decisions in order to achieve desired productivity and product quality. This article is protected by copyright. All rights reserved.

Shedding Light on Data Reconciliation Techniques Applied to Analytical Chemistry

Historically, owing to the increase in processing capacity over the years, validation and adjustment of measurements have become imperative. In particular, concerning discussions related to data and results in analytical chemistry, there is always a need to improve their reliability. The data reconciliation technique has the objective of using measurement redundancies to obtain the best estimate of the true value and, consequently to minimize its uncertainty. Unfortunately, this powerful tool is less known and used by analytical chemists compared to other areas. This approach can be satisfactorily performed in decision-making procedures that focus on chemical analysis, chemometrics, biochemistry analysis, forensics, and environmental sciences, such as in a characterization study, regarding conformance or nonconformance with the specification, doubts related to the malfunctioning of meters and about the compatibility of test methods. This work discusses and sheds light on the importance of data reconciliation, including data reconciliation statistics and application of the technique, traditional data reconciliation in analytical chemistry, principal component analysis based on data reconciliation in analytical chemistry, and fuzzy data reconciliation in analytical chemistry.

Dielectric Spectroscopy to Improve the Production of rAAV Used in Gene Therapy

The insect cell-baculovirus expression vector system is an established method for large scale recombinant adeno-associated virus (rAAV) production, largely due to its scalability and high volumetric productivities. During rAAV production it is critical to monitor process parameters such as Spodoptera frugiperda (Sf9) cell concentration, infection timing, and cell harvest viabilities since they can have a significant influence on rAAV productivity and product quality. Herein we developed the use of dielectric spectroscopy as a process analytical technology (PAT) tool used to continuously monitor the production of rAAV in 2 L stirred tank bioreactors, achieving enhanced control over the production process. This study resulted in improved manufacturing robustness through continuous monitoring of cell culture parameters, eliminating sampling needs, increasing the accuracy of infection timing, and reliably estimating the time of harvest. To increase the accuracy of baculovirus infection timing, the cell growth/permittivity model was coupled to a feedback loop with real-time monitoring. This system was able to predict baculovirus infection timing up to 24 h in advance for greatly improved accuracy of infection and ensuring consistent high rAAV productivities. Furthermore, predictive models were developed based on the dielectric measurements of the culture. These multiple linear regression-based models resulted in correlation coefficients (Q2) of 0.89 for viable cell concentration, 0.97 for viability, and 0.92 for cell diameter. Finally, models were developed to predict rAAV titer providing the capability to distinguish in real time between high and low titer production batches.

A robust flow cytometry-based biomass monitoring tool enables rapid at-line characterization of S. cerevisiae physiology during continuous bioprocessing of spent sulfite liquor

Assessment of viable biomass is challenging in bioprocesses involving complex media with distinct biomass and media particle populations. Biomass monitoring in these circumstances usually requires elaborate offline methods or sophisticated inline sensors. Reliable monitoring tools in an at-line capacity represent a promising alternative but are still scarce to date. In this study, a flow cytometry-based method for biomass monitoring in spent sulfite liquor medium as feedstock for second generation bioethanol production with yeast was developed. The method is capable of (i) yeast cell quantification against medium background, (ii) determination of yeast viability, and (iii) assessment of yeast physiology though morphological analysis of the budding division process. Thus, enhanced insight into physiology and morphology is provided which is not accessible through common online and offline biomass monitoring methods. To demonstrate the capabilities of this method, firstly, a continuous ethanol fermentation process of Saccharomyces cerevisiae with filtered and unfiltered spent sulfite liquor media was analyzed. Subsequently, at-line process monitoring of viability in a retentostat cultivation was conducted. The obtained information was used for a simple control based on addition of essential nutrients in relation to viability. Thereby, inter-dependencies between nutrient supply, physiology, and specific ethanol productivity that are essential for process design could be illuminated.

Multivariate data analysis of capacitance frequency scanning for online monitoring of viable cell concentrations in small-scale bioreactors

Viable cell concentration (VCC) is one of the most important process attributes during mammalian cell cultivations. Current state-of-the-art measurements of VCC comprise offline methods which do not allow for continuous process data. According to the FDA's process analytical technology initiative, process monitoring and control should be applied to gain process understanding and to ensure high product quality. In this work, the use of an inline capacitance probe to monitor online VCCs of a mammalian CHO cell culture process in small-scale bioreactors (250 mL) was investigated. Capacitance sensors using single frequency are increasingly common for biomass monitoring. However, the single-frequency signal corresponds to the cell polarization that represents the viable cell volume. Therefore single-frequency measurements are dependent on cell diameter changes. Measuring the capacitance across various frequencies (frequency scanning) can provide information about the VCC and cope with changing cell diameter. Applying multivariate data analysis on the frequency scanning data successfully enabled direct online monitoring of VCCs in this study. The multivariate model was trained with data from 5 standard cultivations. The model provided a prediction of VCCs with relative errors from 5.5 to 11%, which is a good agreement with the acceptance criterion based on the offline reference method accuracy (approximately 10% relative error) and strongly improved compared with single-frequency results (16 to 23% relative error). Furthermore, robustness trials were conducted to demonstrate the model's predictive ability under challenging conditions. The process deviations in regard to dilution steps and feed variations were detected immediately in the online prediction of the VCC with relative errors between 6.7 and 13.2%. Thus in summary, the presented method on capacitance frequency scanning demonstrates its suitability for process monitoring and control that can save batches, time, and cost. Graphical abstract.

Monitoring online biomass with a capacitance sensor during scale-up of industrially relevant CHO cell culture fed-batch processes in single-use bioreactors

In 2004, the FDA published a guideline to implement process analytical technologies (PAT) in biopharmaceutical processes for process monitoring to gain process understanding and for the control of important process parameters. Viable cell concentration (VCC) is one of the most important key performance indicator (KPI) during mammalian cell cultivation processes. Commonly, this is measured offline. In this work, we demonstrated the comparability and scalability of linear regression models derived from online capacitance measurements. The linear regressions were used to predict the VCC and other familiar offline biomass indicators, like the viable cell volume (VCV) and the wet cell weight (WCW), in two different industrially relevant CHO cell culture processes (Process A and Process B). Therefore, different single-use bioreactor scales (50–2000 L) were used to prove feasibility and scalability of the in-line sensor integration. Coefficient of determinations of 0.79 for Process A and 0.99 for Process B for the WCW were achieved. The VCV was described with high coefficients of determination of 0.96 (Process A) and 0.98 (Process B), respectively. In agreement with other work from the literature, the VCC was only described within the exponential growth phase, but resulting in excellent coefficients of determination of 0.99 (Process A) and 0.96 (Process B), respectively. Monitoring these KPIs online using linear regression models appeared to be scale-independent, enabled deeper process understanding (e.g. here demonstrated in monitoring, the feeding profile) and showed the potential of this method for process control.

Real-time monitoring of ethanol production during Pichia stipitis NRRL Y-7124 alcoholic fermentation using transflection near infrared spectroscopy

The application of in situ near-infrared spectroscopy monitoring of xylose metabolizing yeast such as Pichia stipitis for ethanol production with semisynthetic media, applying chemometrics, was investigated. During the process in a bioreactor, biomass, glucose, xylose, ethanol, acetic acid, and glycerol determinations were performed by a transflection probe immersed in the culture broth and connected to a near-infrared process analyzer. Wavelength windows in near-infrared spectra recorded between 800 and 2200 nm were pretreated using Savitzky-Golay smoothing, second derivative and multiplicative scattering correction in order to perform a partial least squares regression and generate the calibration models. These calibration models were tested by external validation (78 samples). Calibration and validation criteria were defined and evaluated in order to generate robust and reliable models for an alcoholic fermentation process matrix. Moreover, regressions coefficients (β) and variable influence in the projection plots were used to assess the results. A novelty is the use of β versus VIP dispersion plots to determine which vectors have more influence on the response in order to improve process comprehension and operability. Validated models were used in a real-time monitoring during P. stipitis NRRL Y7124 semisynthetic media fermentations.