Cell culture is a complex and involved process, with successful operation contingent upon balancing of numerous critical process parameters to maintain ideal process conditions for optimal cell growth, maintenance, and target product production. Viable cell density (VCD) is a key performance indicator (KPI) of the impact of critical process parameters on the operational performance of the culture. VCD trends are often directly correlated to both product quality and yield, with VCD metrics commonly used to dictate timing of specific process actions such as seed transfer, inoculation, nutrient feeding, and culture harvest to ensure optimal culture titer or productivity. Following the FDA-directed need for improved Process Analytic Technologies (PAT), traditionally off-line VCD measurements are starting to be phased out by continuous in-line measurement technologies.
Measurement of VCD has historically been performed exclusively off-line using cell- staining techniques coupled with imaging analysis or flow cytometry to quantify the concentration of live cells within a representative grab sample. The oldest and most commonly utilized technique for offline measurement of VCD is Trypan blue cell staining. Trypan blue counting utilizes membrane exclusion staining coupled with an image-based method to count the number of viable cells and stained dead cells. The simplicity in method and quantification strategy make Trypan blue a low complexity solution for obtaining VCD counts. However, Trypan blue along with other off-line measurement technologies suffer from limitations that impede complete implementation of recent FDA Process Analytical Technology (PAT) guidelines, specifically labor-intensive manual efforts, limited frequency of sampling, and impact of sampling on culture sterility.
To minimize usage of offline sampling methodologies in-line technologies to measure real-time VCD in process are used. These technologies provide more data with less manual intervention while maintaining culture sterility. Multiple in-line sensor technologies now exist for real-time in-line VCD quantification, each with different measurement principles and generated data.
Examples of In-line Technologies Include:
Imaging technologies such as the Ovizio Imaging System utilize 3D digital holographic microscopy to quantify VCD via phase changes of light passing through cells. The technology has the advantage of being “chemically label-free” in determination of viable cells compared to off-line imaging approaches and can provide additional information on the metabolic state of cells without the need to rely on chemical marker efficacy (Pais, Galrão, et al. 2020). Digital holographic microscopy is limited by the need for advanced algorithms for data dissemination and interpretation.
In-situ spectroscopy such as Raman or fluorescence spectroscopy utilize laser excitation in process to estimate VCD quantifications as well as capture significant information on chemical makeup in-culture (Abu-Absi et al. 2011; Pais et al. 2019). While providing significant in-process data, spectroscopy techniques suffer from calibration and training dataset dependence and require rigorous analysis to generate actionable datasets.
Biocapacitance sensors rely on the principle of dielectric spectroscopy, measuring the storage of electric charge, or capacitance, derived from polarization of live cells within a culture. When applied to cells, this storage of electric charge is also known as biocapacitance. A measure of the permittivity of the cells within a culture is used for an estimation of overall viable cell concentration, while frequency-dependent cell response within the dielectric field can capture cell-specific shifts in physiology that lead to changes in cell biovolume. Biovolume differs from viable cell concentration or cell count in that it refers to the total volume of viable cells, i.e. larger cells make up a proportionally larger part of the signal than the same number of smaller cells. Although straightforward in measurement principle and data output, biocapacitance directly measures viable cell biovolume. This can be correlated to a known VCD reference to account for cell geometry if a cell count is the desired output. With distinct advantages and disadvantages in complexity and data generated for each in-line sensor technology, it can be difficult to choose the best technology for a process. The following sections present a case for utilizing biocapacitance as a robust and easily implementable VCD measurement technology, with or without other measurement tools, and explains when and why biocapacitance-derived VCD measurements differs from VCD values derived from offline measurements. It should be understood that when the results of the measurements differ it is because they are measuring different aspects of the same culture. This will be discussed in more detail in the following sections.
Permittivity-Based Viable Cell Density for Cell Culture Monitoring
In-line permittivity sensors are generally straightforward to implement within a cell culture. The measurement technology does not require process calibration or training datasets. The measurement is not affected by changes in e.g. media contributions, and typically needs minimal data processing or complex modeling to generate a VCD measurement. This makes permittivity measurement an ideal in-line measurement for adaptation to legacy processes or applications validated around specific off-line measurement numbers, requiring minimal effort to correlate permittivity measurements to off-line VCD values. Permittivity measurements are compatible with different process scales, with permittivity-derived metrics at one scale being directly transferrable with scale-up or scale-down activities.
To date, permittivity has been successfully implemented in numerous cell lines and applications. For examples of common cell types and applications, download the Hamilton Cell Density eBook
When Off-line VCD and In-line Permittivity Do Not Correlate
When implementing in-line permittivity in an established process, comparing the permittivity signal to historical off-line VCD values is an important step towards using permittivity signal as a proxy for a continuous viable cell count. In many cases permittivity signal trends will correlate linearly to off-line VCD metrics; however, there are instances where the difference in measuring techniques between permittivity measurements and offline VCD will result in varying degrees of measurement correlation.
Cell Counts (Off-line)
Measurement Technique Differences
High Cell Concentrations
At high cell concentrations (commonly considered > 1×107 cells/mL in mammalian cultures) off-line VCD samples must be diluted before measurement, leading to increased sampling error (Cadena-Herrera et al. 2015). Additionally, high cell concentrations are often associated with cell aggregation, with cell clumping impacting offline image-based cell counting algorithms. In-line permittivity sensors do not require dilution, and the measurement is minimally impacted by cell aggregation, resulting in lower measurement errors compared to off-line VCD measurements.
Adherent Cell Cultures
Off-line VCD measurement commonly requires the cells to be detached from microcarriers or other matrices prior to analysis. The need for detachment presents steps where incomplete detachment and partial reattachment can occur, skewing VCD counts. Trypsin, commonly used for cell detachment, also poses toxicity impacts on cells that are repeatedly exposed or exposed for prolonged periods of time. In-line permittivity measurement is not impacted by most commercially available adherent substrates and is capable of providing a direct measurement of viable cell density of an adherent cell line without the need for cell detachment and exposure to toxic chemicals (Justice et al. 2011).
Irregular Cell Geometries
While many industrially relevant cell lines have a roughly spherical shape, there are also cell lines with more irregular cell geometries. These cell lines are harder to quantify effectively with off-line measurements as cell heterogeneity affects the consistency with which an offline image-based measurement can resolve individual cells on a 2D image. Permittivity continuously measures the biovolume of cells regardless of geometry, with continuous measurements averaging out the influence of cell geometry to provide a robust aggregate VCD measurement.
Quantifying Cell Viability
A key feature of cell apoptosis is loss of membrane integrity that in the end leads to membrane leakage. Off-line cytometry or imaging approaches rely upon the use of molecular dyes or specific binding tags to either physically penetrate the damaged membrane or bind to specific membrane components that become accessible during varying stages of apoptosis.
For instance, Annexin V binds to phosphatidylserine, a membrane-bound protein that becomes exposed during early stages of cellular apoptosis, while caspase assays quantify the presence of caspase, an enzyme typically formed intracellularly and secreted during mid-stage cell apoptosis. Trypan Blue only becomes effective for quantifying cell viability during late stage apoptosis, where the physical degradation of the membrane allows for dye penetration and staining of the non-viable cell.
Permittivity measurement is based upon intracellular ion separation and is sensitive to cellular changes in cytoplasmic conductivity such as a decrease in intracellular conductivity due to active transport failure in the cell membrane. As a result, permittivity detects loss of viability at an earlier stage of apoptosis compared to many staining and molecular binding offline techniques (A. Henslee et al. 2016).
Processes with Changes in Cell Biovolume and Membrane Properties
Although highly correlated to viable cell density counts (often in cells/mL and also known as viable cell concentration) in most applications, the measurement of permittivity is fundamentally a measurement of viable cell biovolume within a cell membrane. It will capture changes in cell biovolume due to both environmental and physiological factors impacting the cell culture.
Environmental stressors such as culture osmolarity impacts the cells’ ability to regulate osmotic balance, swelling, or shrinking depending upon ionic concentration strength of the suspension. Nutrient limitations may also impact cell size, promoting a transition in metabolic stage and resulting in a change in culture population cell size (Pan et al. 2017). Cell size may also vary with respect to the application or target product. For instance, viral vector applications have noted cell swelling and altered membrane properties in response to transfection and intracellular viral production. In each of these cases, a permittivity sensor will measure these biovolume increases, even if the number of viable cells is unchanged. As a result, the permittivity signal becomes both a quantification of viable cell concentration and cell physiology shift in-process. Permittivity measurement can be used effectively with or without other metrics of VCD measurement. Understanding the impact of measuring principles on quantified values and how measurement technologies differ allows for informed decisions for utilizing each technology most effectively.
Permittivity Response and the Cell Cycle
Permittivity for Process Insight
As discussed in the previous section, a continuous biocapacitance-based measurement can generate unique culture information that cannot be obtained via typical offline analysis. In practical application of dielectric spectroscopy, a biocapacitance probe generates an entire frequency response distribution of cells within a dielectric field, referred to as the beta dispersion curve. Multivariate parameters determined through empirical fit of the beta dispersion curve to the Cole-Cole model for a spherical particle in a dielectric field help characterize changes in the beta dispersion curve shape and magnitude (Dabros et al. 2009). The usefulness of the multivariate parameters becomes apparent when trying to capture shifting cellular properties in real time. Used as an actionable process control metric, permittivity has practical applications beyond monitoring and indication of cell culture trends and VCD. The following are three examples of applications in which the unique benefits of permittivity measurement were utilized to enhance processes in ways that other measurement types could not.
Viral Vector Production
Viral vector production is an application associated with fairly drastic changes in cell physiology arising from morphology changes during viral accumulation and viral secretion mechanisms that impact cell membrane properties. Permittivity has been used successfully to capture the distinct phases of viral component synthesis and release in both insect (Petiot et al. 2017), and mammalian host systems (Ansorge et al. 2011). A direct application of successful viral transfection culture characterization allowed for determination of high versus low performing cultures, as well as a prediction of time for optimal culture harvest using the multivariate parameters determined from permittivity measurement (Pais, Brown, et al. 2020).
Identifying Nutrient Limitations
Nutrient limitations have also been successfully observed using biocapacitance through changing fscan and multivariate parameters. By providing a continuous measurement of both permittivity and accompanying multivariate parameters, detection of changes to culture health in the form of decreased VCD can be readily identified with a continuous permittivity measurement. As an example, once a subtle decrease in VCD profile was identified, investigation of discrepancies in the permittivity signal from historical growth curves highlighted a nutrient deficiency impacting culture growth. Subsequent modification of bolus addition with deficient nutrients added at observed time points improved both the maximum VCD generated during growth phase and expanded the viable production phase (Ansorge, Esteban, and Schmid 2010). Multivariate parameters were also shown to respond uniquely to nutrient limitation, demonstrating an ability to capture changes in intracellular properties of the cell in response a shifting cellular metabolism (Ansorge et al. 2010).
If captured early enough within a process, correction of a nutrient limitation can mitigate onset of apoptosis within the culture. Multivariate properties of permittivity signal have been used to identify early onset apoptosis, allow time for process correction, and subsequently act as an analogous indicator of return to normal cellular physiology. As a result, proactive monitoring of multivariate properties of cells within a dielectric field can be useful not only to generate a correlative metric to VCD, but to capture actionable cellular mechanistic changes such as apoptosis or metabolic shifts within a continuous measurement fingerprint (Ma et al. 2019).
Modified Cell Culture Calculations and Feeding Strategies
Replacement of fixed time bolus feeding has been successfully implemented with permittivity as the calculable parameter for the integral of viable cell concentration, or area under the growth curve. In this strategy, permittivity replaces the offline VCD measurement for calculation of total culture growth (Moore, Sanford, and Zhang 2019; Fernandes et al. 2019). Permittivity signal presents a unique advantage to feed control compared to offline VCD measurement in the form of capturing both changes in cell concentration and changes in cell size. Depending upon the stage of cell lifecycle and stage of the culture (growth or stationary) cell size may differ, with larger cells requiring more nutrients for equivalent cell metabolism. Feeding based upon offline VCD measurement alone may not be sufficient for effective nutrient dosing in these cases (Downey et al. 2014).
The need for effective PAT in the form of process monitoring, characterization, and control is a growing topic of discussion within biopharma community. Biocapacitance is an important tool in reaching this goal. This in-line and continuous VCD measurement tool is a mature technology with a proven track record of success in a multitude of cell culture applications and will continue to expand in utilization capacity as time progresses.
1. Abu-Absi, Nicholas R., Brian M. Kenty, Maryann Ehly Cuellar, Michael C. Borys, Sivakesava Sakhamuri, David J. Strachan, Michael C. Hausladen, and Zheng Jian Li. 2011. “Real Time Monitoring of Multiple Parameters in Mammalian Cell Culture Bioreactors Using an In-Line Raman Spectroscopy Probe.” Biotechnology and Bioengineering 108 (5): 1215–21. https://doi.org/10.1002/bit.23023.
2. A. Henslee, Erin, Ruth M. Torcal Serrano, Fatima H. Labeed, Rita I. Jabr, Christopher H. Fry, Michael P. Hughes, and Kai F. Hoettges. 2016. “Accurate Quantification of Apoptosis Progression and Toxicity Using a Dielectrophoretic Approach.” Analyst 141 (23): 6408–15. https://doi.org/10.1039/C6AN01596D.
3. Ansorge, Sven, Geoffrey Esteban, and Georg Schmid. 2010. “On-Line Monitoring of Responses to Nutrient Feed Additions by Multi-Frequency Permittivity Measurements in Fed-Batch Cultivations of CHO Cells.” Cytotechnology 62 (2): 121–32. https://doi.org/10.1007/s10616-010-9267-z.
4. Ansorge, Sven, Olivier Henry, Marc Aucoin, Robert Voyer, John P. Carvell, and Amine Kamen. 2010. “Monitoring the Cell Size Distribution of Mammalian Cell Cultures Using On-Line Capacitance Measurements.” In Cells and Culture, edited by Thomas Noll, 853–59. Dordrecht: Springer Netherlands. https://doi.org/10.1007/978-90-481-3419-9_150.
5. Ansorge, Sven, Stéphane Lanthier, Julia Transfiguracion, Olivier Henry, and Amine Kamen. 2011. “Monitoring Lentiviral Vector Production Kinetics Using Online Permittivity Measurements.” Biochemical Engineering Journal 54 (1): 16–25. https://doi.org/10.1016/j.bej.2011.01.002.
6. Cadena-Herrera, Daniela, Joshua E. Esparza-De Lara, Nancy D. Ramírez-Ibañez, Carlos A. López-Morales, Néstor O. Pérez, Luis F. Flores-Ortiz, and Emilio Medina-Rivero. 2015. “Validation of Three Viable-Cell Counting Methods: Manual, Semi-Automated, and Automated.” Biotechnology Reports 7 (September): 9–16. https://doi.org/10.1016/j.btre.2015.04.004.
7. Dabros, Michal, Danielle Dennewald, David J. Currie, Mark H. Lee, Robert W. Todd, Ian W. Marison, and Urs von Stockar. 2009. “Cole–Cole, Linear and Multivariate Modeling of Capacitance Data for on-Line Monitoring of Biomass.” Bioprocess and Biosystems Engineering 32 (2): 161–73. https://doi.org/10.1007/s00449-008-0234-4.
8. Downey, Brandon J., Lisa J. Graham, Jeffrey F. Breit, and Nathaniel K. Glutting. 2014. “A Novel Approach for Using Dielectric Spectroscopy to Predict Viable Cell Volume (VCV) in Early Process Development.” Biotechnology Progress 30 (2): 479–87. https://doi.org/10.1002/btpr.1845.
9. Fernandes, Juhi, Jayme Currie, Kevin Ramer, and An Zhang. 2019. “Development of Capacitance Tools: At-Line Method for Assessing Biomass of Mammalian Cell Culture and Fixed Cell Calibration Standard.” Biotechnology Journal 14 (4): e1800283. https://doi.org/10.1002/biot.201800283.
10. Justice, C., J. Leber, D. Freimark, P. Pino Grace, M. Kraume, and P. Czermak. 2011. “Online- and Offline- Monitoring of Stem Cell Expansion on Microcarrier.” Cytotechnology 63 (4): 325–35. https://doi.org/10.1007/s10616-011-9359-4.
11. Ma, Fuduo, An Zhang, David Chang, Orlin D. Velev, Kelly Wiltberger, and Rashmi Kshirsagar. 2019. “Real-Time Monitoring and Control of CHO Cell Apoptosis by in Situ Multifrequency Scanning Dielectric Spectroscopy.” Process Biochemistry 80 (May): 138–45. https://doi.org/10.1016/j.procbio.2019.02.017.
12. Moore, Brandon, Ryan Sanford, and An Zhang. 2019. “Case Study: The Characterization and Implementation of Dielectric Spectroscopy (Biocapacitance) for Process Control in a Commercial GMP CHO Manufacturing Process.” Biotechnology Progress 35 (3): e2782. https://doi.org/10.1002/btpr.2782.
13. Pais, Daniel A. M., Chris Brown, Anastasia Neuman, Krishanu Mathur, Inês A. Isidro, Paula M. Alves, and Peter G. Slade. 2020. “Dielectric Spectroscopy to Improve the Production of RAAV Used in Gene Therapy.” Processes 8 (11): 1456. https://doi.org/10.3390/pr8111456.
14. Pais, Daniel A. M., Paulo Galrão, Anastasiya Kryzhanska, Jérémie Barbau, Inês A. Isidro, and Paula Alves. 2020. “Holographic Imaging of Insect Cell Cultures: Online Non-Invasive Monitoring of Adeno-Associated Virus Production and Cell Concentration.” Processes 8 (4): 487. https://doi.org/10.3390/pr8040487.
15. Pais, Daniel A. M., Rui M. C. Portela, Manuel J. T. Carrondo, Inês A. Isidro, and Paula M. Alves. 2019. “Enabling PAT in Insect Cell Bioprocesses: In Situ Monitoring of Recombinant Adeno‐associated Virus Production by Fluorescence Spectroscopy.” Biotechnology and Bioengineering 116 (11): 2803–14. https://doi.org/10.1002/bit.27117.
16. Pan, Xiao, Ciska Dalm, René H. Wijffels, and Dirk E. Martens. 2017. “Metabolic Characterization of a CHO Cell Size Increase Phase in Fed-Batch Cultures.” Applied Microbiology and Biotechnology 101 (22): 8101–13. https://doi.org/10.1007/s00253-017-8531-y.
17. Petiot, Emma, Sven Ansorge, Manuel Rosa-Calatrava, and Amine Kamen. 2017. “Critical Phases of Viral Production Processes Monitored by Capacitance.” Journal of Biotechnology 242 (January): 19–29. https://doi.org/10.1016/j.jbiotec.2016.11.010.
Download this Article in PDF Format
In-line vs Off-line: Permittivity Measurements versus Viable Cell Counts in Cell Culture Monitoring
Cell culture is a complex and involved process. Learn how successful operation is contingent upon balancing numerous critical process parameters to maintain ideal process conditions for optimal cell growth, maintenance, and target product production.
Download the In-line vs Off-line Article
Fill out the form below to download a PDF format of this article, In-line vs Off-line: Permittivity Measurements versus Viable Cell Counts in Cell Culture Monitoring.