System and method for real-time prediction of biomass of cells cultured in fixed bed bioreactor

Abstract

A method of monitoring biomass during cell culture of cells in a bioreactor is provided. The method comprises: culturing the cells in the bioreactor using a cell culture medium perfused through the bioreactor; measuring at least one of cellular nutrients and cellular byproducts in the cell culture medium; determining at least one of a rate of depletion of the cellular nutrients and a rate of accumulation of the cellular byproducts; and predicting the number of cells within the bioreactor at a specified culture time based on at least one of the depletion rate and the accumulation rate.

Description

[0001] Cross-references to related applications

[0002] This application claims the priority of U.S. Provisional Application Serial No. 63 / 546,657, filed October 31, 2023, and U.S. Provisional Application Serial No. 63 / 524,279, filed June 30, 2023, pursuant to 35 USC §119, the contents of which are incorporated herein by reference in their entirety. Technical Field

[0003] This disclosure generally relates to systems and methods for monitoring and predicting cell culture in bioreactor systems. Specifically, this disclosure relates to methods, protocols, systems, and models for monitoring biomass in cell culture within bioreactor systems and for using oxygen uptake rate (OUR) to simulate cell attachment and growth. Background Technology

[0004] In the bioprocessing industry, large-scale cell culture is required to produce hormones, enzymes, antibodies, vaccines, therapeutic proteins, and cell therapies. The cell and gene therapy market is growing rapidly, with promising treatments entering clinical trials and quickly moving towards commercialization. However, a single cell therapy session may require billions of cells or trillions of viruses. Therefore, the ability to provide large quantities of cell products in a short period is crucial for clinical success.

[0005] Most cells used in bioprocessing are adhesion-dependent, meaning they need to adhere to a surface to grow and function. Traditionally, adherent cell culture is carried out on two-dimensional (2D) cell attachment surfaces incorporating one of a variety of container forms, such as T-flasks, Petri dishes, cell factories, cell stack containers, roller bottles, and other multilayer containers (e.g., HYPERStack® from Corning Inc.). These methods can have significant drawbacks, including difficulty in achieving sufficiently high cell densities for large-scale production of therapeutic agents or cells.

[0006] Alternative methods have been proposed to increase the volumetric density of cultured cells. These methods include microcarrier cultures performed in stirred tanks; hollow fiber bioreactors, in which cells proliferate within the fibrous spaces between the spaces, forming larger three-dimensional aggregates; and packed-bed bioreactors. In packed-bed or fixed-bed bioreactors, a filled or fixed cell substrate is used to provide an attachment surface for adherent cells. Culture medium is perfused along the surface or through the semi-porous substrate to provide the nutrients and oxygen required for cell growth. For example, packed-bed bioreactor systems containing a support or matrix system for encapsulating cells have been previously disclosed in U.S. Patents 4,833,083, 5,501,971, and 5,510,262. The packed-bed matrix is ​​typically made of porous particles or polymer nonwoven microfibers as the substrate.

[0007] One of the significant problems with traditional fixed-bed bioreactors is the non-uniformity of cell distribution within the bed. For example, a packed bed can act as a depth filter, where cells are primarily trapped in the inlet region or other areas with relatively low flow rates and / or high substrate density, resulting in a gradient distribution of cells during the inoculation step. Furthermore, due to random fiber packing, the flow resistance and cell retention efficiency across the packed bed cross-section are uneven. For instance, the culture medium flows rapidly through areas with low cell packing density and slowly through areas with higher resistance due to a greater number of trapped cells. This creates a channeling effect, where nutrients and oxygen are delivered more efficiently to areas with lower volumetric cell density, while areas with higher cell density remain under suboptimal culture conditions.

[0008] Another significant drawback of conventional packed-bed systems disclosed in the prior art is the inability to efficiently collect intact, live cells at the end of the culture process. Cell collection is crucial if the end product is cells, or if the bioreactor is used as part of a “seed train” where cell populations are grown in one container and then transferred to another for further population growth. U.S. Patent No. 9,273,278 discloses a bioreactor design for improving the efficiency of cell recovery from a packed bed during the cell collection step. This design is based on relaxing the packed bed matrix and agitating or stirring the packed bed particles to allow the porous matrix to collide, and thus detach the cells. However, this approach is time-consuming and laborious and can lead to significant cell damage, thereby reducing overall cell viability.

[0009] Furthermore, due to the random arrangement of fibers in conventional packed or fixed-bed substrates, bioreactor users may find it difficult to predict cell culture performance, as substrate arrangement and / or packing varies between cultures. Monitoring the health or progress of cell cultures is also challenging. For example, the very presence of a fixed bed limits the options for effectively monitoring culture health and biomass production. Moreover, the packed substrates of conventional fixed-bed bioreactors make efficient cell collection extremely difficult or impossible because cells are believed to be trapped in the packed bed, further hindering the understanding of cell culture performance.

[0010] Regardless of the platform used, the early stages of process development require users to have information to better understand cell behavior, virus production, and culture progress. Upstream bioreactor process development necessitates identifying key parameters and quality characteristics, as well as parameter definitions and their relationship to the end product. Understanding what these parameters are and how they can scale with higher densities or larger systems is crucial for process development and efficiency.

[0011] Upstream bioprocess production also adheres to Good Manufacturing Practice (GMP) regulations and requirements involving process analytical technology (PAT). PAT is considered a tool for the design, analysis, and control of production processes. The quality of the final product can be ensured by measuring process parameters and product characteristics. This can include extensive online monitoring of the culture process, providing a useful tool for process characterization and the detection of process changes. Relevant parameters for process characterization and control in packed bed bioreactors include pH, temperature, dissolved oxygen or oxygen delivery (DO2), and carbon dioxide (CO2). However, one of the identified drawbacks of packed bed bioreactors is the difficulty in obtaining substrate samples to directly assess cell state and overall cell culture progress. Obtaining substrate samples carries the risk of contaminating the entire culture or, in the case of a non-uniform platform, providing misleading or inaccurate data.

[0012] Biomass is a crucial measure in cell culture operations and is one of the most desirable productivity and product quality indicators for large-scale mammalian cell culture and biomanufacturing. Real-time biomass monitoring is essential for assessing process quality, improving productivity, and developing model-based process controls such as feed planning, perfusion rate control, and transfection and cell collection timing.

[0013] In recent years, fixed-bed or packed-bed bioreactors have been increasingly used for the scale-up production of cells, viral vectors, extracellular vesicles, therapeutic proteins, and potentially cultured meat. Recently, Corning® commercialized the Ascent® fixed-bed bioreactor system, which can be scaled up to large-scale bioreactor systems for the biomanufacturing market. For example, this fixed-bed system can be available in different sizes (defined by the effective surface area for cell attachment), ranging from 1 m². 2 2.5 m 2 and 5 m 2 (This can be considered as "process development" or "PD" scale) up to 20 m 2 50 m 2 and 100 m 2 (This can be considered a "pilot" scale), up to 200 m 2 500 m 2 and 1000 m 2 (This can be viewed as "production" scale). Scalability is a key aspect of advancing a process from the development phase to production scale. The ability to monitor all relevant parameters with the same measurement types across every process scale is highly desirable to maintain high product quality and quantity and to comply with GMP standards. The United States Food and Drug Administration (FDA) recommends using process analysis techniques as a mechanism for designing, analyzing, and controlling pharmaceutical manufacturing processes by measuring key process parameters that affect critical quality attributes.

[0014] Fixed-bed bioreactors are increasingly used for scaling up production in adherent cell cultures. When cell culture is involved, biomass monitoring is a crucial tool for designing, analyzing, and controlling pharmaceutical manufacturing processes. For suspension cell cultures, biomass monitoring can be achieved using optical or electrical methods. However, for adherent cell cultures using fixed-bed bioreactors, no validated methods or sensors are available for online biomass monitoring. Historically, substrate (e.g., glucose) consumption rates have been proposed as a means of understanding cell growth kinetics. However, because the glucose consumption rate (GCR) is influenced by many factors beyond the number of viable cells, there are no established protocols, methods, or models for obtaining and using effective substrate consumption rates and metabolite accumulation rates for biomass calculation and prediction.

[0015] While it is possible to manufacture viral vectors for early clinical trials using existing platforms, a platform capable of producing high-quality products in larger quantities is needed to achieve later-stage commercial manufacturing scale. Additionally, systems and methods are required to collect specific measurable parameters from cell cultures during bioreactor operation to better control all aspects of the culture process in real time and to detect and diagnose abnormal culture conditions. Summary of the Invention

[0016] According to embodiments of this disclosure, a method for monitoring the biomass of cells during cell culture in a bioreactor is provided. The method includes: culturing the cells in the bioreactor using a cell culture medium perfused through the bioreactor; measuring at least one of cell nutrients and cell byproducts in the cell culture medium; determining at least one of the consumption rate of the cell nutrients and the accumulation rate of the cell byproducts; and predicting the number of cells in the bioreactor at a specified culture time based on at least one of the consumption rate and the accumulation rate. According to one aspect of the embodiment, the bioreactor is a fixed-bed bioreactor having a substrate configured to culture cells attached to the surface of the substrate. At least one cell nutrient may be glucose or glutamine; and at least one cell byproduct may be lactate or ammonia. According to another aspect of the embodiment, the cell culture medium may be a glucose- or glutamine-rich cell culture medium.

[0017] According to various aspects of the embodiments, measuring at least one of the cell nutrients and the cell by-products in the cell culture medium includes performing multiple measurements of the cell nutrients or the cell by-products, the interval between the multiple measurements being less than the doubling time of the cells in the cell culture. The measurement interval is greater than or equal to a minimum interval time, the minimum interval time being the time during which the change in the level of the cell nutrients or the cell by-products exceeds the measurement tolerance for measuring the cell nutrients or the cell by-products. In various aspects of the embodiments, the minimum interval time is greater than or equal to about 30 minutes, 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours, 11 hours, 12 hours, 13 hours, 14 hours, 15 hours, 16 hours, 17 hours, 18 hours, 19 hours, or 20 hours. Measuring the cell nutrients in the cell culture medium may include measuring the cell nutrients multiple times daily in the cell culture. Measuring the cell by-products in the cell culture medium may include measuring the cell by-products multiple times daily in the cell culture.

[0018] According to various aspects of the embodiments, predicting the cell number includes calculating the biomass at the specified culture time using a mathematical model. The measurement may include using an inline sensor in the perfusion line of the cell culture medium. The measurement may also include offline measurement using a sample of the cell culture medium.

[0019] According to various aspects of the embodiments, the method further includes, after determining the consumption rate and the accumulation rate, comparing at least one of the consumption rate of a first cell nutrient and the accumulation rate of a first cell byproduct with at least one of the consumption rate of a second cell nutrient and the accumulation rate of a second cell byproduct. The comparison may include comparing the consumption rate of glucose with the consumption rate of glutamine, comparing the consumption rate of glucose with the accumulation rate of ammonia, comparing the accumulation rate of lactate with the consumption rate of glutamine, and / or comparing the accumulation rate of lactate with the accumulation rate of ammonia. The method may further include determining anomalies in the cell culture based on the comparison. The method may further include inoculating cells in the bioreactor at a certain inoculation density. The method may further include supplying fresh cell culture medium to the bioreactor, and the measurement may include a first measurement performed at least one hour after supplying the fresh cell culture medium. According to various aspects of the embodiments, the cell number N is predicted. t This includes using the following equation:

[0020] N t = N 接种 e kt

[0021] Where N 接种 The number of cells inoculated into the bioreactor is given by k, the cell growth rate is given by t, and the time for predicting the number of cells is given by t.

[0022] According to embodiments of this disclosure, a method for real-time biomass prediction in a cell culture bioreactor is provided. According to aspects of these embodiments, the method may include measuring glucose concentration at predetermined time intervals. The method further includes selecting at least one mathematical model for biomass prediction based on at least one of glucose measurement frequency, availability of maximum achievable cell density, and a predetermined lag time of the cell line cultured in the bioreactor. The method includes fitting the measured glucose concentration according to the at least one mathematical model. The method may further include predicting biomass in real time after a period of time during which cells have been cultured beyond the predetermined lag time of the cultured cell line. Attached Figure Description

[0023] Figure 1This is a schematic illustration of a cell culture system according to one or more embodiments.

[0024] Figure 2 The operation of controlling the perfusion flow rate of a cell culture system is demonstrated according to one or more embodiments.

[0025] Figure 3 For use according to one or more embodiments Figure 1 A diagram showing the bioreactor infusion flow rate and oxygen concentration over time during operation of an exemplary bioreactor system.

[0026] Figure 4A In order to be in Figure 3 A graph showing the dissolved oxygen concentration over time during the operation of the bioreactor.

[0027] Figure 4B In order to be in Figure 3 A graph showing the pH over time during the operation of the bioreactor.

[0028] Figure 4C In order to be in Figure 3 A graph showing the media conditioning temperature over time during the operation of a bioreactor.

[0029] Figure 5 In order to be in Figure 3 A graph showing the oxygen consumption of packed bed cell cultures over time during bioreactor operation, including the slope α of the curve.

[0030] Figure 6 This is a diagram showing the slope α of a bioreactor relative to cell seeding density according to one or more embodiments.

[0031] Figure 7 This is a diagram showing the slope α relative to the bottom surface area of ​​the cell culture medium according to one or more embodiments.

[0032] Figure 8 A graph showing the cell collection yield relative to the slope α according to one or more embodiments.

[0033] Figure 9 A graph illustrating the correlation between the actual glucose curve and the model fitted line according to the embodiment.

[0034] Figure 10 A graph illustrating the correlation between the actual lactate curve and the model fitted line according to the embodiment.

[0035] Figure 11 This is a diagram comparing the actual total number of cells with the predicted number at the time of collection, according to an embodiment.

[0036] Figure 12 This is to illustrate the glucose curve according to the embodiment and the graph of data fitting using a modified exponential growth model.

[0037] Figure 13A A diagram illustrating the data fitting results using the Logistic model according to an embodiment.

[0038] Figure 13B A diagram illustrating the data fitting results using the Gompertz model according to an embodiment.

[0039] Figure 13C A diagram illustrating the data fitting results using the Stannard model according to an embodiment.

[0040] Figure 14A To demonstrate the situation at 2.5 m according to the embodiment 2 The inoculation density in the bioreactor was 22 K / cm². 2 In the HEK239TAAV2 bioproduction experiment, the DO curves of the medium leaving the bioreactor and the flow rate of the medium were obtained from the MCV.

[0041] Figure 14B To demonstrate the situation at 2.5 m according to the embodiment 2 The inoculation density in the bioreactor is 35 K / cm². 2 In the HEK293 cell culture experiment, the DO curves in the medium leaving the bioreactor and the flow rate of the medium were plotted.

[0042] Figure 15A According to the embodiments Figure 9 A graph of oxygen uptake rate (OUR) over time in the HEK293T AAV2 bioproduction experiment of A.

[0043] Figure 15B According to the embodiments Figure 9 A graph of oxygen uptake rate (OUR) over time in HEK293 cell culture experiments of B.

[0044] Figure 16A According to the embodiment, the oxygen consumption rate q is fitted. o2 The obtained diagram of cell attachment dynamics.

[0045] Figure 16B According to the embodiment, the oxygen consumption rate q is fitted. o2 The obtained diagram of cell attachment dynamics.

[0046] Figure 16CAccording to the embodiment, the oxygen consumption rate q is fitted. o2 The obtained diagram of cell attachment dynamics.

[0047] Figure 16D This is a graph showing the change over time of the percentage of cells found in a bioreactor using an offline cell counting protocol with a single-phase model according to an embodiment.

[0048] Figure 17 A diagram illustrating the OUR curves and data fitted using the logistic cell growth model according to the embodiment.

[0049] Figure 18 A diagram illustrating the OUR curves and data fitted using the logistic cell growth model according to the embodiment.

[0050] Figure 19 This is a diagram illustrating the change in cell number over time using real-time biomass prediction according to an embodiment. Detailed Implementation

[0051] Various embodiments of this disclosure will be described in detail with reference to the accompanying drawings (if any). The various embodiments mentioned do not limit the scope of the invention, which is limited only by the scope of the appended claims. Furthermore, any examples set forth in this specification are not limiting and are merely examples of many possible embodiments of the claimed invention.

[0052] Embodiments of this disclosure include systems and methods for monitoring and controlling cell cultures. This disclosure describes systems and methods for collecting specific signal signatures during bioreactor operation to enable better real-time control of critical aspects and to detect and diagnose abnormal culture conditions. The identified signature parameters of cell cultures described in this disclosure can be used as tools for implementing process analysis techniques and online monitoring of upstream processes. Therefore, optimized cell culture production processes can be established by developing routines and reproducibility for the signature operating parameters.

[0053] According to embodiments of this disclosure, bioreactor systems and methods are provided for monitoring the status of cell cultures in a bioreactor system during cell culture operations. Specifically, embodiments describe bioreactor systems with an outlet sensor at the outlet of a cell culture bioreactor or vessel, systems capable of real-time signal collection and processing of signals from this sensor and / or other sensors, and cell culture methods using said systems. For example, methods include using the sensor signals as trigger points for important cell culture process steps, or for predicting the expected health status of cell cultures over time for a particular bioreactor size or inoculation density, or assessing their current health status. Advantages of these systems and methods include the ability to proactively monitor bioreactor status in real time without requiring physical sampling of the packed bed substrate for offline analysis. Continuous monitoring of bioreactor status will also allow end-users to proactively adjust bioprocess steps depending on the progress of the culture process within the packed bed bioreactor. The ability to characterize and record the progress of a bioprocess operation will further allow end-users to monitor and record batch-to-batch consistency of the process. This type of tracking of the progress and consistency of cell culture operations can be highly advantageous.

[0054] In conventional large-scale cell culture bioreactors, different types of packed-bed bioreactors have been used. Typically, these packed beds contain a porous matrix to retain adherent or suspension cells and support growth and proliferation. The packed-bed matrix provides a high surface area to volume ratio, thus allowing for higher cell densities than in other systems. However, packed beds often act as depth filters, where cells are physically trapped or entangled in the fibers of the matrix. Therefore, because the cell inoculum flows linearly through the packed bed, cells experience a non-uniform distribution within the bed, resulting in variations in cell density across the depth or width of the bed. For example, the cell density may be higher at the inlet region of the bioreactor, while significantly lower closer to the outlet. In another instance, non-uniformity in the packed bed creates a channeling effect, where cell culture medium preferentially flows in certain areas of the bed while being restricted to other areas, similarly leading to non-uniform cell distribution and non-uniform or inconsistent distribution of culture medium or nutrients. This uneven distribution of cells within the packed bed severely hinders the scalability and predictability of such bioreactors in bioprocess manufacturing, and may even lead to reduced efficiency of cell growth or viral vector production per unit surface area or volume of packed bed.

[0055] Another problem encountered with existing packed-bed bioreactors is the channeling effect described above. Due to the randomness of the nonwoven fibers used in the packing, the local fiber density is non-uniform at any given cross-section of the packed bed. The culture medium flows faster in areas of low fiber density (higher bed permeability) and much slower in areas of high fiber density (lower bed permeability). This resulting non-uniform perfusion of the culture medium across the packed bed creates a channeling effect, which manifests itself as a significant nutrient and metabolite gradient, negatively impacting overall cell culture and bioreactor performance. Cells located in areas with low culture medium perfusion will starve and often die due to nutrient deficiency or metabolite poisoning. Cell collection is another problem encountered when using bioreactors filled with nonwoven fiber scaffolds. Because the packed bed acts as a deep filter, cells released at the end of the cell culture process are trapped within the packed bed, resulting in extremely low cell recovery rates. This severely limits the use of such bioreactors in bioprocesses using live cells as products. Therefore, non-uniformity leads to different regions being exposed to different flow rates and shear forces, which significantly reduces the available cell culture area, causes uneven culture, and interferes with transfection efficiency and cell release.

[0056] To address these and other problems with existing cell culture protocols, embodiments of this disclosure provide bioreactor systems, cell growth substrates, matrices of said substrates, and methods of using said bioreactor systems and substrates, enabling efficient and high-yield cell culture of adherent cells and production of cell products (e.g., proteins, antibodies, viral particles). Embodiments include porous cell culture substrates made of ordered and regular arrays of porous substrate materials, enabling uniform cell seeding and culture medium / nutrient perfusion, as well as efficient cell collection. Embodiments also enable scalable cell culture protocols where the substrate and bioreactor can perform cell seeding and growth and / or cell product collection from process development scale to full production scale without sacrificing the uniformity of the embodiments. For example, in some embodiments, the bioreactor can be readily scaled from process development scale to production scale, with a viral genome quantity (VG / cm³) per unit substrate surface area throughout the production scale. 2 The collectability and scalability of the embodiments described herein enable them to be used for efficient seed amplification to grow cell populations at multiple scales on the same cell substrate. Furthermore, the embodiments described herein provide a cell culture substrate with a high surface area, which, in combination with the other features described, enables high-yield cell culture protocols. In some embodiments, for example, the cell culture substrate and / or bioreactor discussed herein can produce 10 [units of something] per batch. 16 Up to 10 18 One viral genome (VG).

[0057] Embodiments of this disclosure can realize a viral vector platform with a practical size, which can be produced in batches of more than about 10. 14 One viral genome, more than approximately 10 per batch 15 One viral genome, more than approximately 10 per batch 16 One viral genome, more than approximately 10 per batch 17 One viral genome or up to or greater than approximately 10 per batch 16 Viral genomes are produced at a scale of approximately 10 viral genomes per batch. In some embodiments, the yield is approximately 10 per batch. 15 To about 10 18 One or more viral genomes. For example, in some embodiments, the viral genome yield can be about 10. 15 To about 10 16 One viral genome or batch, or approximately 10 per batch 16 To about 10 19 One viral genome, or approximately 10 per batch 16 Up to 10 18 One viral genome, or approximately 10 per batch 17 To about 10 19 One viral genome, or approximately 10 per batch 18 To about 10 19 One viral genome, or approximately 10 per batch 18 One or more viral genomes.

[0058] Furthermore, the embodiments disclosed herein not only enable cells to attach and grow on a cell culture substrate, but also enable the collection of live cultured cells. The inability to collect live cells is a significant drawback of current platforms, making it difficult to build and maintain a sufficient number of cells to achieve production capacity. According to one aspect of embodiments of this disclosure, live cells can be collected from a cell culture substrate, comprising between 80% and 100% live cells, or between about 85% and about 99% live cells, or between about 90% and about 99% live cells. For example, in the collected cells, at least 80% are live cells, at least 85% are live cells, at least 90% are live cells, at least 91% are live cells, at least 92% are live cells, at least 93% are live cells, at least 94% are live cells, at least 95% are live cells, at least 96% are live cells, at least 97% are live cells, at least 98% are live cells, or at least 99% are live cells. Cells can be released from the cell culture substrate using, for example, trypsin, TrypLE, or Accutase.

[0059] According to one or more embodiments, a cell culture bioreactor may include a cell culture substrate within a bioreactor vessel. The substrate may be deployed in a filled-bed bioreactor configuration or in other configurations within a three-dimensional culture chamber of the bioreactor vessel. Due to contamination concerns, the vessel may be a disposable container that can be discarded after use.

[0060] like Figure 1 As shown, embodiments of this disclosure include a bioreactor system 100 for culturing cells in a cell culture vessel 100. The cell culture vessel includes an inlet 112 and an outlet 114 fluidly connected to an internal reservoir 111 of the cell culture vessel 110. The internal reservoir 111 contains space for containing and culturing cells and may also include a cell culture substrate (not shown) on which adherent cells can be cultured. In some embodiments, the inlet 112 is located at one end of the cell culture vessel 110 for introducing culture medium, cells, and / or nutrients into the cell culture vessel 110, and the outlet 114 is located at the opposite end for removing culture medium, cells, or cell products from the cell culture vessel 110. The substrate within the internal reservoir can take many forms, some of which are illustrated herein. Some embodiments may use one or both of the inlet 112 and outlet 114 to allow culture medium, cells, or other contents to flow into and out of the cell culture vessel 110. For example, inlet 112 can be used to allow culture medium or cells to flow into cell culture vessel 110 during cell seeding, perfusion, and / or culture phases, but can also be used to remove one or more of culture medium, cells, or cell products through inlet 112 during collection phases. Therefore, the terms “inlet” and “outlet” are not intended to limit the function of these openings, but should generally be understood to mean ports used for introducing and removing fluid, respectively, during routine processes of cell growth. Outlet sensor 118 is located at outlet 114 of cell culture vessel 110. As used herein, “at outlet” can mean a sensor located inline on the fluid flow path that receives culture medium from outlet 114 and returns it to another part of the system (e.g., a culture medium conditioning container), or it can mean a sensor located within cell culture vessel 110, but preferably after the cell culture substrate, packed bed, or other cell culture zone within cell culture vessel 110. In this way, outlet sensor 114 can detect the characteristics of the culture medium after it has passed through the packed bed, cell culture substrate, or other cell culture zone.

[0061] The system further includes a culture medium conditioning vessel (MCV) 120 capable of containing and conditioning cell culture medium 122. Fluid flow paths 142, 144 deliver conditioned culture medium 112 from the MCV 120 to the cell culture vessel 110 and return used culture medium from the cell culture vessel 110 to the MCV 120. The MCV 120 may be coupled to a plurality of sensors and / or conditioning components 124a, 124b, 124c, 124d for sensing the characteristics of the cell culture medium and adjusting or conditioning the medium as needed during cell culture. These include, but are not limited to, dissolved gas (e.g., O2, air, CO2, N2) sensors and supplies, pH sensors, oxygenators / gas bubbling units, temperature probes and temperature control devices, and nutrient and substrate addition ports. The gas mixture supplied to the bubbling unit may be controlled by gas flow controllers for N2, O2, and CO2 gases. The culture medium conditioning vessel 120 may also contain an impeller for mixing the culture medium.

[0062] The system may further include a culture medium conditioning control unit 130 operatively connected to a plurality of sensors and / or conditioning components 124a, 124b, 124c, 124d to process signals detected from these sensors and / or control the conditioning components to regulate the culture medium 122 within the MCV 120. The culture medium conditioning control unit 130 may also be operatively connected to a pump 150 to control the pump 150 and thus control the rate of fluid flow through fluid flow paths 142, 144 and the rate of perfusion through the cell culture vessel 110. Alternatively, the pump 150 may be directly connected to the outlet sensor 118 or connected via a perfusion control unit separate from the culture medium conditioning control unit 130. In some embodiments herein, a peristaltic pump is used, but other types of pumps may also be used. Figure 1 As shown, the culture medium conditioning container 120 is provided as a separate container from the bioreactor container 110. This allows for the separation of culture medium conditioning from cell culture, with the conditioned medium then supplied to the cell culture space. However, in some embodiments, culture medium conditioning may be performed within the bioreactor container 110.

[0063] In some embodiments, the culture medium conditioning control unit 130 can be used to maintain stable or desired levels of various parameters in the cell culture medium 122 within the MCV 120, thereby maintaining the host culture medium 122 at specific temperatures, oxygen saturation levels, pH, and CO2 concentrations. For example, for a given cell line or stage of a cell culture process, the cell culture medium 122 may require specific temperatures, pH levels, dissolved gas content, or nutrient levels to achieve optimal cell health and / or growth. Culture medium 122 from the culture medium conditioning container 120 is delivered to the cell culture container 110 through an inlet 112, which may also include an injection port for inoculating cell seed material and initiating cell culture. The cell culture container 110 may also include an outlet 114 through which the cell culture medium 122 exits the container 110. Additionally, cells or cell products may be discharged through the outlet 114. An outlet sensor 118 is provided for analyzing the composition of the effluent from the cell culture container 110. As described above, the culture medium regulation control unit 130 can receive a signal from the outlet sensor 118 (e.g., an O2 sensor) and, based on said signal, adjust the fluid flow through the cell culture container 110 by sending a signal to a pump 150 (e.g., a peristaltic pump) upstream of the inlet 112 of the cell culture container 110. Therefore, based on one or a combination of factors measured by the outlet sensor 118, the pump 150 can control the flow into the cell culture container 110 to obtain the desired cell culture conditions. Because the cell culture medium 122 within the MCV 120 can be maintained in the desired state, changes in flow rate can effectively address any needs of the cells within the cell culture container 110. For example, because the cell culture medium 122 exiting the MCV 120 is regulated for optimal performance, the culture medium entering through the inlet 112 should meet the optimal requirements of the culture medium. If the outlet sensor 118 detects a less than ideal level of the cell culture medium leaving the cell culture vessel 110 at the outlet 114, this could mean, for example, that the cells in the culture have consumed a certain amount of dissolved gases (e.g., oxygen) or cell nutrients from the medium, and that at least some cells (i.e., those near the outlet where the most medium has been consumed) are not being cultured optimally. Therefore, for example, if the dissolved oxygen level in the cell culture medium at the outlet sensor 118 is below the optimal level (e.g., for a given cell type, culture stage, etc.), the perfusion flow rate can be increased to supply a higher rate of regulated medium, which should then allow all cells (even those near the outlet) to be cultured under optimal conditions.

[0064] The culture medium perfusion rate is controlled by a culture medium conditioning control unit 130, which collects and compares sensor signals from sensors 124a-124d in the culture medium conditioning vessel 120 and MCV 120, as well as the outlet sensor 118. Due to the fluidity of the culture medium perfusion through the packed bed substrate in the cell culture vessel 110, a gradient of nutrients, pH, and oxygen is formed along the packed bed. The perfusion flow rate of the bioreactor can be automatically controlled by the culture medium conditioning control unit 130, which is operatively connected to the pump 150. This control scheme... Figure 2 This is shown in the flowchart. Figure 2 In the sensing and control process 200 shown, in step 202, optimal conditions are pre-determined through a round of bioreactor optimization operation. These optimal conditions include minimum pH, minimum oxygen level, and nutrient (e.g., glucose) at outlet sensor 118, and pH, oxygen level, and nutrient (e.g., glucose) in MCV 120. These parameters are provided as examples, and those skilled in the art will understand that other parameters (e.g., temperature) may also be relevant to a given application. The pH and oxygen levels in MCV 120 are controlled independently based on inputs from the corresponding sensors located in MCV 120. The nutrient (e.g., glucose) level in MCV 120 is maintained in part based on the signal from outlet sensor 118, such that the nutrient level in MCV 120 remains higher than the nutrient level detected by outlet sensor 118. Steps 204 and 206 are performed in parallel during cell culture operation. In step 204, outlet sensor 118 is used to measure conditions (e.g., pH, O2, and glucose) at outlet 114 of cell culture vessel 110. In step 206, sensors 124a-124d are used to measure conditions in MCV 120 (e.g., pH, O2, and glucose). In step 208, the infusion pump 150 is controlled by the control unit based on inputs from both steps 202 and 204. In step 210, it is determined whether: the pH at outlet sensor 118 is greater than the minimum pH determined in step 202; whether the oxygen at outlet sensor 118 is greater than the minimum oxygen level determined in step 202; and whether the nutrient level in MCV 120 is greater than the nutrient level at outlet sensor 118, and whether the nutrient level at the outlet sensor is greater than the minimum level determined in step 202. If all these conditions are met, the pump continues infusion at the current flow rate (step 212). If these conditions are not met, step 214 queries whether the current infusion rate is less than or equal to the maximum flow rate. If not, the system reassesses the minimum pH, O2, and glucose at outlet 114, or increases the nutrient level in MCV 120. However, if the current perfusion rate is less than the maximum flow rate, step 218 instructs an increase in the perfusion flow rate. Sensing and control scheme 200 returns to...Figure 2 Steps 204 and 206 are performed at the top of the chart.

[0065] According to embodiments of this disclosure, nutrient and / or oxygen consumption by cells within the cell culture vessel can therefore be directly measured and responded to in a manner that maintains the desired conditions for the cells. For example, during cell culture operations, the culture medium regulation control unit 130 is pre-programmed to maintain a specific level of oxygen saturation in the bulk culture medium volume relative to atmospheric saturation, wherein said level in the MCV is measured by sensors 124a-124d. A second sensor (outlet sensor 118) is placed at the bioreactor outlet 114 to measure the oxygen saturation level in the culture medium immediately after it leaves the cell culture vessel. Using these sensors and controllers, a constant oxygen consumption level can be maintained within physiological conditions by automatically adjusting the perfusion flow rate.

[0066] According to some embodiments of this disclosure, systems and methods for improved process monitoring are provided, which can accelerate process development of cell culture protocols and improve the efficiency and reproducibility of cell culture processes. The ability to characterize and record the progress of these bioprocess operations will allow end-users to monitor and record batch-to-batch consistency of the process. Relevant parameters for process characterization include cell growth, cell quality, culture medium conditions (temperature, pH, pO2, and pCO2), and metabolite concentrations (glucose, lactate, glutamine, and ammonium). While the temperature, pH, pO2, and pCO2 of the host culture medium are routinely controlled online in cell culture, these and other process parameters are not currently monitored online in dynamic systems. Therefore, embodiments of this disclosure provide systems and methods for obtaining, for example, oxygen consumption parameters in a packed-bed perfusion bioreactor, and demonstrate that such parameters are characteristic of a given bioprocess and can therefore be used as marker parameters for that bioprocess.

[0067] As discussed above, Figure 1A schematic diagram of a bioreactor system (e.g., a packed-bed perfusion bioreactor) is presented. The cell culture medium entering the cell culture vessel 110 through inlet 112 can have 100% atmospheric oxygen saturation. Alternatively, according to Henry's law, the concentration of a gas in the liquid phase is equal to the Henry's law constant (k) multiplied by the partial pressure of the gas in the gas phase; therefore, oxygen saturation can be expressed as the oxygen concentration in the cell culture medium, and at standard atmospheric pressure, 100% saturation equals 204 μM. During the passage of the culture medium through the cell culture vessel, dissolved oxygen is used by the immobilized cells, and its concentration in the cell culture medium decreases. Different cell types have different oxygen consumption rates. However, a bioreactor system with the sensing and control system of this disclosure allows the user to operate the process with a specified oxygen concentration at the bioreactor outlet, measured by outlet sensor 118, and the culture medium conditioning and perfusion control system according to... Figure 2 The logical operations presented in the flowchart.

[0068] To illustrate this sensing and control in bioreactor systems, some examples will be presented. Specifically, Figure 3 A typical graph is shown showing the percentage of dissolved oxygen (302) over time during bioreactor operation, as measured by outlet sensor 118, and the corresponding perfusion rate (304) (ml / min) of the culture medium in the system. The flow rate is automatically controlled by a peristaltic perfusion flow control unit. In this example, the bioreactor is inoculated with cells at time 0:00, and the user sets the minimum oxygen saturation level of the culture medium at outlet sensor 118 to 30%. The initial culture medium perfusion flow rate is set to 33 ml / min. The inoculated cells are then fed into the bioreactor system and begin to attach to the packed bed substrate and proliferate. As a result, oxygen consumption increases, and the saturation level of the culture medium at the outlet decreases to approximately 30% at 26 hours post-inoculation. Therefore, the control system automatically increases the perfusion flow rate to maintain a minimum oxygen saturation level of 30% at bioreactor outlet 114. At 72 hours post-inoculation, the user decreases the minimum outlet oxygen saturation level setting from 30% to 15%, and cell culture proceeds in automatic mode. It should be noted that the culture medium conditions are maintained independently by a culture medium conditioning control unit in the culture medium conditioning vessel. An example of adjusting the parameters of a culture medium container is shown in Figure 4A , 4B And presented in 4C. Specifically, Figure 4A The percentage of oxygen in the MCV medium over time is shown. Figure 4B The pH of the culture medium over time was shown, and Figure 4C The temperature of the culture medium over time is shown.

[0069] As described above, the embodiments include real-time signal processing and control of the bioreactor system, as well as the development of characteristic signal flags for specific bioreactor operations, which can be used as analytical tools to compare and validate independent bioreactor operations. Therefore, characteristic signal flags can be used to assess the health status of cells cultured within cell culture vessels and to make decisions regarding subsequent process steps occurring during bioreactor operation. For example, as discussed above, Figure 3 The recorded oxygen saturation concentration at bioreactor outlet 114 over time is shown during the cell culture process. The oxygen concentration level at the bioreactor outlet decreased from approximately 82% at time point 0 to approximately 30% during the first 26 hours of bioreactor operation. While this decrease occurred, the oxygen concentration at bioreactor inlet 112 remained constant, as shown in the diagram. Figure 4A The values ​​detected in MCV 120 are shown. Using the oxygen concentrations at the inlet and outlet, the rate of oxygen consumption of cells in the culture can be determined using Equation 1:

[0070] .

[0071] Equation (1)

[0072] Figure 5 The oxygen consumption rate over time (in hours) is shown (expressed as as% / min). Figure 5 It also shows a dashed line representing the approximate slope α of the line, which can be used as a characteristic signal indicator of bioreactor operation. In other words, Figure 5 The slope α in the data directly reflects the cell culture progress within the bioreactor system. This value can be used as a process analysis tool to control and describe upstream bioprocesses. The following example illustrates this, from a system similar to... Figure 5 The slope α of the graph is directly related to the health of the cell culture and the biomass within the packed bed substrate.

[0073] To illustrate the use of parameter α, multiple cell culture runs were performed using bioreactor systems with different sizes of cell culture substrates, bed heights, and cell seeding densities. Table 1 summarizes the parameters used in the seven cell culture runs.

[0074]

[0075] Table 1 An overview of the operation of the seven bioreactor systems, including bed height (cm), total number of inoculated cells (millions of cells), and total bed surface area (cm²). 2 Seeding density (cells / cm²) 2 ), Total number of cells collected (billion cells), viability of collected cells (%), collection density (cells / cm³)2 ), maximum perfusion rate (ml / min), maximum O2 consumption (AU), and slope α.

[0076] As shown in Table 1, multiple bioreactors were inoculated with varying numbers of cells, ranging from 151 to 453 million cells per bioreactor. Three identical bioreactors (Bioreactors #1, #2, and #3) had the same bed height (2.7 cm), were inoculated with the same number of cells (151 million cells per bioreactor), and had the same total bed surface area (6780 cm²). 2 ) and seeding density (22,222 cells / cm²) 2 The other three identical bioreactors (#4, #5, and #6) had the same packed bed height (5.4 cm), were inoculated with the same number of cells (302 million cells per bioreactor), and had the same total packed bed surface area (13,560 cm²). 2 ) and seeding density (22,227 cells / cm²) 2 The final bioreactor (#7) features an increased bed height (8.1 cm), a total number of inoculated cells (453 million cells), and a filled bed surface area (20,340 cm²). 2 However, it has a similar seeding density (22,222 cells / cm²). 2 During the five-day culture process, the host culture medium conditions (pH, DO2, temperature, and CO2) are maintained in automatic mode by a control system according to one or more embodiments described herein. The control system operates the culture medium conditioning vessel to maintain the culture medium conditions, wherein... Figures 4A-4C This represents typical measurement results for controlled culture media. The culture medium perfusion flow rate of the bioreactor system is automatically maintained to keep the DO2 at the bioreactor outlet at a specific saturation level. Again, Figure 3 The diagrams shown represent typical perfusion flow rates and culture medium outlet DO2 observed during these experiments. From, for example... Figure 3 The graph in the diagram derives the value of total oxygen consumption, similar to... Figure 5 As shown in the figure. For each bioreactor run, the slope of the linear curve fit of these patterns was determined (e.g., Figure 5 The slope α, as determined above, is presented in the last column of Table 1. The value of slope α can be used as a process analysis tool to control and describe upstream bioprocesses and predict biomass yield within the packed bed substrate.

[0077] For example, Figure 6 , 7Figures 8 and 9 respectively plot the α values ​​from Table 1 against the number of seeded cells, the surface area of ​​the packed bed, and the collection density. The linearity of these plots can be used to predict cell culture responses based on various cell culture system parameters. For example, Figure 6 The linearity of the graphs in Table 1 indicates that the upstream processes developed for the small-scale bioreactors #1 and #2 in Table 1 can be scaled up by 2x and 3x, respectively, for bioreactors #4 through #7. Therefore, continuous monitoring and recording of the slope α value can be used to determine the scalability of the upstream process. An alternative way to verify process scalability is to plot the slope α relative to the surface area of ​​the bioreactor, as shown in the figure. Figure 7 As shown. Figure 6 and 7 The orange data points correspond to the failed bioreactor #3 from Table 1 (discussed below).

[0078] Monitoring the slope α value during bioreactor operation serves as a characteristic signal marker reflecting the health and expansion of cell culture. For example, as shown in Table 1, bioreactors #1, #2, and #3 were inoculated with the same number of cells. The characteristic signal marker (slope α) was measured for all bioreactors. Figure 8 This indicates that real-time monitoring of the slope α can be used to compare the performance of identical bioreactors and predict bioreactor productivity. Figure 8 It can be seen that bioreactor #3 is operating under suboptimal conditions that result in the lowest cell yield. Therefore, monitoring the value of the slope α during bioprocess operation can serve as a characteristic signal of a given process, and any process deviations can be detected if the predetermined value is not within the range defined during process development and optimization.

[0079] According to some embodiments, the culture medium conditioning vessel is controlled by a controller to provide appropriate temperature, pH, O2, and nutrients. While in some embodiments the bioreactor may also be controlled by the same controller, in other embodiments the bioreactor is positioned in a separate perfusion loop where a pump is used to control the flow rate of the culture medium through the perfusion loop based on the detection of O2 at or near the bioreactor outlet.

[0080] The foregoing describes systems and methods for biomass calculation and prediction based on exponential growth models, as well as related biomass prediction schemes and methods. Embodiments of this disclosure also include several alternative mathematical models and related methods and schemes for predicting the biomass of adherent cell cultures in a fixed-bed bioreactor with improved accuracy in real time. These mathematical models were developed for modeling bacterial growth curves. The growth kinetics of mammalian cells during culture are much slower than their rapid growth kinetics. These mathematical models have been used to understand the growth behavior of mammalian cells, but have not yet been used to predict biomass based on the kinetic characteristics of metabolites in the cell culture medium. Various aspects of the embodiments compare the biomass prediction accuracy of different mathematical models and define the boundary conditions for each model to accurately predict biomass. Therefore, embodiments of this disclosure include several different mathematical models for predicting the biomass of cell cultures in a fixed-bed bioreactor and compare their accuracy in predicting biomass.

[0081] In the embodiments, when glucose and lactate are analyzed at high frequencies (e.g., every 1 minute, every 5 minutes, every 10 minutes, every 30 minutes using online sensors), biomass can be accurately predicted in real time using exponential growth models, modified exponential growth models, modified Monod's models, logistic models, Gompertz models, and Steinard models.

[0082] In the embodiments, when only one substrate (e.g., glucose, glutamine) or one metabolite (e.g., lactate, ammonia) is analyzed at high frequencies (e.g., every 1 minute, every 5 minutes, every 10 minutes, every 30 minutes using an online sensor, an in-line sensor, or an at-line sensor), the exponential growth model, the modified exponential growth model, the logistic model, the Gompertz model, and the Steinard model can be used to accurately predict biomass in real time.

[0083] In an embodiment, when metabolites are analyzed at low frequencies (e.g., every 30 minutes, every 2 hours, every 4 hours, every 8 days, or twice a day using an online sensor, in-line sensor, near-line sensor, or offline analyzer), biomass can be accurately predicted using an exponential growth model or a modified exponential growth model.

[0084] In embodiments, when metabolites are analyzed at low frequencies (e.g., every 30 minutes, every 2 hours, every 4 hours, every 8 days, or twice daily using online sensors, in-line sensors, near-line sensors, or offline analyzers) and the maximum cell density and typical lag time of the cultured cell lines are predetermined, biomass can be accurately predicted using logistic, Gompertz, and Steinard models by estimating the initial values ​​of the maximum cell density and lag time. The maximum cell density and typical lag time of the cell lines can be determined experimentally.

[0085] Embodiments of this disclosure include a method for real-time prediction of biomass in a cell culture bioreactor. According to aspects of these embodiments, the method may include measuring glucose concentration at predetermined time intervals. The method further includes selecting at least one mathematical model for biomass prediction based on at least one of glucose measurement frequency, availability of maximum achievable cell density, and a predetermined lag time of the cell line cultured in the bioreactor. The method includes fitting the measured glucose concentration to the at least one mathematical model. The method may further include real-time prediction of biomass after a period of time during which cells have been cultured beyond the predetermined lag time of the cultured cell line.

[0086] Experimental Section

[0087] The data used for modeling were obtained using a fixed-bed bioreactor made of vertically stacked polyethylene terephthalate (PET) mesh reels, with a total surface area of ​​approximately 1 m². 2 Culture medium samples were collected from the medium conditioning container (MCV) every 30 minutes using a 1 mL syringe for 8 hours daily. Each morning, half the culture medium in the MCV was drained and replaced with the same volume plus 40 mL of additional medium to contain the previous day's samples in fresh, warm medium. All collected samples were analyzed offline using a Nova Biomedical Flex II analyzer. HEK293T cells were used. The cell seeding density was 22,000 cells / cm². 2 Cell attachment was found to be completed within 3.5 hours, with >95% of cells attaching.

[0088] Modified Monod Model

[0089] The predictive accuracy of the modified Mono model was evaluated here. The modified Mono model assumes that a specific amount of nutrient (or substrate) can produce a certain amount of biomass, and that the cell growth rate is determined by the actual concentration of a specific substrate (e.g., glucose). For mammalian cells, glucose is the primary substrate driving cell growth, while lactate, the main metabolite of glucose, tends to inhibit cell growth when lactate accumulates to elevated levels after cells reach high density.

[0090] Based on the modified Monod model that takes into account the growth-limiting effect of glucose and the inhibitory effect of lactate, the overall glucose uptake rate (or the glucose consumption rate per cell) is calculated. for

[0091]

[0092] in It is the maximum glucose consumption rate per cell [mmol G / 10] 8 [cells / day], It is the half-rate constant of glucose (i.e., the glucose concentration at which the glucose consumption rate is 50% of its maximum consumption rate). It is the half-rate constant of lactate (i.e., the lactate concentration at which the lactate accumulation rate is 50% of its maximum accumulation rate), C G It is the glucose concentration, and C L It refers to the lactate concentration.

[0093] This model shows that biomass is proportional to nutrients, therefore:

[0094] Equation (2)

[0095] Equation (3)

[0096] Equation (4)

[0097] in It refers to biomass (or cell number). It is the amount of glucose produced per cell

[10] 8 [cells / mMG], It is the lactate produced per unit of glucose consumed [mmol L / mmol G]. Figure 9 and 10 It shows glucose ( Figure 9 ) and lactate ( Figure 10 The original concentration curve of ) is closely fitted with equations (2)-(4).

[0098] K is obtained by fitting the original data with a modified Mono model.G 2.5 g / L The concentration was 25.3 g / L, indicating that in this culture experiment, if the concentrations of glucose and lactate were relatively low, glucose exhibited a relatively small cell growth limiting effect (especially in the later stages of culture), while the inhibitory effect of lactate was small or negligible.

[0099] The data fitting also yielded an estimated cell number of 2.25E+09 compared to the actual cell number at collection time of 1.79E+09, indicating that the modified Mono model tends to lead to a significant overestimation of biomass.

[0100] To further explore the requirements of data quality for prediction accuracy, the same experimental data were analyzed, but the frequency of metabolite measurements was artificially reduced—that is, the data points were artificially reduced to measurements every 1 hour, every 2 hours, every 4 hours, and every 8 hours, instead of measurements every 30 minutes. The results showed that as the frequency of data decreased, the modified Mono model generally led to increasingly poorer accuracy in biomass prediction (i.e., increasingly overestimation of biomass) (see Table 2).

[0101]

[0102] Table 2. Comparison of predicted cell numbers using the modified Mono model with analyses of glucose and lactate concentrations at different frequencies.

[0103] These results indicate that, for high accuracy in biomass prediction, the modified Mono model requires high-quality data, and preferably, real-time measurement of glucose and lactate. In other words, when real-time online measurement of glucose and lactate is available, the modified Mono model can be applied for real-time biomass prediction.

[0104] Modified Exponential Model

[0105] The above disclosure provides an exponential growth model and related methods and schemes for biomass prediction. This simple exponential growth model assumes that glucose consumption per cell is constant and that nutrient concentrations rarely fall below a threshold concentration, which may cause some growth restriction but will not cause cells to deviate from their growth path.

[0106] This model has proven effective and accurate in biomass prediction, with accuracy within ±15% for almost all culture experiments across a wide range of cell lines (e.g., HEK293, HEK293T, CHO, CHO-5 / 9α) across various bioreactor sizes, under several culture medium treatments and metabolite sampling frequencies. The accurate predictive power of the simple exponential growth model stems from the fact that cells never reach a stagnant phase at the end of any of these bioreactor culture experiments and continue to follow their growth pathways throughout the culture. These results also indicate that a minimum of twice-daily substrate sampling is required for accurate biomass prediction using this simple model, and online real-time monitoring is not necessary.

[0107] Figure 11 This demonstrates the use of a range of bioreactor sizes (0.5 m) under several culture medium treatment protocols and glucose sampling frequencies. 2 1 m 2 2.5 m 2 50 m 2 The actual total number of cells and the predicted number of cells at collection were compared for different cell culture experiments conducted herein. The predicted cell number was obtained using the simple exponential growth model disclosed herein.

[0108] However, during cell culture experiments, nutrients (especially glucose) are likely to drop to such low concentrations over extended periods that they could cause significant growth restriction, while the cells may still maintain their growth pathway. It is noteworthy that the effect of lactate can be neglected because lactate concentrations never reach levels that would have a significant inhibitory effect on cell growth under perfusion culture conditions in fixed-bed bioreactors (such as the Corning Ascent system). To address these situations and further improve predictive accuracy, a modified exponential growth model is introduced here. Here, the glucose consumption rate r per cell can be varied according to the actual glucose concentration C:

[0109]

[0110] It is the maximum glucose consumption rate. It is the half-rate constant of glucose (i.e., the glucose concentration at which the glucose consumption rate is 50% of its maximum consumption rate).

[0111] Therefore, for a bioreactor, the total glucose consumption rate is

[0112]

[0113] Where V is the volume of the culture medium, Q is the flow rate of the culture medium, and C... 进C is the glucose concentration when the culture medium enters the bioreactor. 出 This is the glucose concentration when the culture medium leaves the bioreactor, and N is the total number of cells.

[0114] After rearranging, it becomes:

[0115]

[0116] After integration, it becomes:

[0117] Equation (5)

[0118] Using equation (5) to fit the same set of dates, the glucose consumption rate was 1.6E-3 mg / min / million cells, the cell growth rate was 3.56E-4, and the predicted cell number was 1.80E+09, indicating that the modified exponential growth model has higher accuracy than the previous simple exponential growth model (Table 3). This improvement in biomass prediction is evidenced by the fact that at the end of day 3, the glucose concentration dropped to approximately 0.5 g / L, i.e., the concentration that induces some growth restriction ( Figure 12 ). Figure 12 The glucose curves according to the examples and the data fitting using a modified exponential growth model are shown. It is worth noting that 2.5 g / L of K was used here. G and 0.0016 mg / min / cell r max Perform data fitting. K G and r max Both values ​​were obtained by fitting the data using the Mono model described above. Both values ​​may be cell line dependent. For different cell types, these values ​​can be obtained using a modified Mono model, following the same protocol and method described above.

[0119]

[0120] Table 3. Comparison of cell numbers predicted using the simple and modified exponential growth models.

[0121] This analysis demonstrates that the modified exponential growth model can be used for accurate biomass prediction even when glucose concentrations unexpectedly fall below the threshold level at which growth restriction begins to be induced. The threshold level is likely cell-dependent; a concentration of approximately 0.5 g / L was found in both HEK293 and HEK293T cells.

[0122] Logistic Model, Gompertz Model and Stannard Model

[0123] The logistic model assumes that cultured cells follow a perfect logistic population growth path, beginning with a lag phase, followed by an exponential growth phase and a stasis phase. According to this model, the number of cells N at a given time is:

[0124] Equation (6)

[0125] Here, a, b, and c are three mathematical parameters, not biologically significant parameters.

[0126] The Gompertz model, an empirical S-shaped model, has been widely used to fit cell growth data (Tjørve KMC, Tjørve E (2017). The use of Gompertz models in growth analyses, and new Gompertz-model approach: An addition to the Unified-Richards family. PLoS ONE 12(6): e0178691). Several reparameterizations of the Gompertz model have been reported in the literature. One type II Gompertz model assumes that a single parameter a controls the initial value of the curve (i.e., the intersection with the y-axis), where both parameters b and c make the initial point a relative value (a percentage of the upper asymptote) and do not affect the initial point. According to this model, the total number of cells N is:

[0127] Equation (7)

[0128] Here, a, b, and c are three mathematical parameters, not biologically significant parameters.

[0129] The Steinard model, another empirical S-shaped model, is described by four free parameters, each influencing the characteristics of the curve: an initial lag or slow growth phase; a rapid exponential growth phase; and a phase of decreasing growth rate. According to this model, the total number of cells N is:

[0130] Equation (8)

[0131] Here, a, p, l, and k are three mathematical parameters, not biologically significant parameters.

[0132] Based on the continuous stirred tank reactor model of an MCV coupled with a bioreactor, the total amount of cell culture substrate consumed during a given time period is a function of the cell consumption rate r and the total cell number N:

[0133] Equation (9)

[0134] By combining equation (9) with equations (6), (7), or (8) and fitting the same experimental data, the corresponding mathematical parameters can be obtained and the cell number at collection time can be predicted. The results are summarized in Table 4. The results show that all three models yielded fairly good predictions, but compared with this 1 m 2 Compared to the maximum cell number achievable by bioreactors (approximately 5.0E+09 for HEK293T cells), the Gompertz and Steinard models appear to derive a much higher maximum cell number.

[0135]

[0136] Table 4. Summary of parameters and cell numbers predicted using three different models.

[0137] Figure 13A , 13B And 13C demonstrates the use of the logistic model ( Figure 13A ), Gumperts model ( Figure 13B ) and the Steinard model ( Figure 13C The data fitting results.

[0138] Similarly, all three S-shaped growth models contain three or four mathematical parameters (a, b, c, l, k, and p). Since these parameters have largely no biological significance, it is difficult to estimate their initial values ​​for an accurate fit. However, all of these growth models can be rewritten using biological parameters A(= ln(N)). ∞ / N0), maximum asymptote), µ max (Maximum growth rate) and λ (lag time or duration of slow growth) were used to replace mathematical parameters (Zwietering, MH et al., Modeling of the bacterial growth curve. Appl. Envrion. Microbiol. 1990, 56, 1875-1881).

[0139] For the logistic model:

[0140] Equation (10)

[0141] For the Gompert model:

[0142] Equation (11)

[0143] For the Steinard model:

[0144] Equation (12)

[0145] Here ν is the shape parameter, and e is exp(1).

[0146] Since equations 10, 11, and 12 all contain biological parameters, initial values ​​can be set for data fitting. For example, when setting the maximum number of cells... Limiting it to 8 billion will delay the time lag or slow growth period. When the time limit was 3 days, fitting the same dataset with these S-shaped models significantly improved the accuracy of biomass prediction (Table 5).

[0147]

[0148] Table 5. Comparison of parameters and cell numbers predicted using three different S-shaped models after providing estimated starting values ​​for maximum cell number and lag time.

[0149] *****

[0150] The preceding text describes various systems and methods for biomass calculation and prediction based on exponential growth models, as well as related biomass prediction schemes and methods. It also presents several alternative mathematical models and related methods and schemes for real-time prediction of biomass from adherent cell cultures in fixed-bed bioreactors with improved accuracy. These systems and methods include those for predicting biomass in 1m... 2A set of parameters and analytes in the culture medium of the human embryonic kidney HEK-293T cell line cultured in an Ascent® (from Corning Incorporated) bioreactor were characterized, and several analytes, including glucose, were identified. The curves of these analytes can be used as a basis for biomass prediction. Several models and methods for accurately predicting biomass in bioreactors of different sizes under a range of culture, culture medium treatments, and analyte sampling conditions have been disclosed above. However, glucose-based models have two significant drawbacks. First, these models may not allow for the estimation of cell attachment kinetics because glucose consumption during cell attachment is minimal. Second, these models may not allow for biomass estimation during cell transfection and subsequent viral vector production, as cell transfection and bioproduction alter the cell's substrate consumption behavior. Currently, cell attachment kinetics are typically estimated using offline cell sampling and counting methods, which are labor-intensive and have relatively low temporal resolution. Similarly, for adherent cultures, biomass is usually measured offline at the end of the experiment because no validated methods or sensors are available for direct online biomass monitoring. Therefore, further embodiments of this disclosure include systems and methods for simulating cell attachment and growth kinetics using oxygen uptake rate (“OUR”). In the examples discussed herein, the OUR value is obtained using an in-line oxygen sensor. Cell attachment and growth kinetics are then estimated by fitting an OUR curve with a corresponding model and comparing it with offline data.

[0151] Experimental Setup

[0152] The bioreactors used in these cell culture experiments included two different sizes (2.5 m). 2 and 5 m 2 The Ascent® bioreactors (from Corning Incorporated) are used. All bioreactors have fixed beds made of polyethylene terephthalate (PET) mesh discs. The diameters of these bioreactors and the number of PET discs stacked vertically within the housing vary. Surface area represents an estimated surface area effective for cell attachment and growth. The mesh is made of PET fibers with a filament diameter of approximately 160 µm and an opening size of approximately 250 µm, an opening area of ​​approximately 37%, and an effective thickness of 274 µm. For a 60 mm disc, the effective surface area of ​​this type of mesh is estimated to be approximately 75 cm². 2 The effective volumetric surface area of ​​all bioreactors is approximately 80 cm². 2 / mL. All bioreactors are pre-sterilized before use.

[0153] Both HEK293T and HEK293 were used in cell culture experiments. HEK293T (a subline of HEK293) contains the gene for the SV40 large T antigen and is commonly used for retrovirus production. Both cell lines were first passaged in T175 flasks. For culture in the bioreactor, a standard, internally developed cell attachment protocol was used. First, the cells were suspended in an optimal volume of culture medium, depending on the size of the bioreactor. Cell seeding densities varied, but were typically around 22,000 cells / cm². 2 Next, the cell suspension was added to the culture medium conditioner vessel (MCV) and then flowed through the bioreactor at the optimal rate until the cells in the MCV were depleted >95%. Cell concentration was quantified offline on the Vi-Cell system every 15 minutes during the first hour and every 30 minutes thereafter. Finally, the flow rate of the cell culture was reduced.

[0154] For cell culture, use an appropriate culture medium replacement protocol to maintain sufficient nutrients to support cell growth. For example, for cell seeding and expansion culture, HEK293-T cells are cultured at 35 K / cm². 2 Inoculated into Ascent® PD 5.0 ​​m 2 The cells were cultured in the reactor for 4 days. Due to the high seeding density and long culture duration, cell culture medium was added twice daily; each time, half the total volume of cell culture medium was replaced. Table 6 below shows the volume and flow rate for cell seeding and culture in bioreactors of different sizes.

[0155]

[0156] Table 6. Culture medium volume and flow rate for cell inoculation and culture in bioreactors of different sizes.

[0157] A pair of oxygen sensors (PreSens, Regensburg, Germany) were used to monitor dissolved oxygen (DO) concentrations in the culture medium conditioning vessel (“MCV”) and immediately removed from the bioreactor vessel, respectively. DO concentrations were reported as a percentage of typical DO levels in the culture medium maintained under standard culture conditions, which was approximately 18.5% (pO2 140 mmHg, 182 µM).

[0158] Cells were transfected using a triple transfection protocol to generate the AAV2 vector carrying the GFP transgene. HEK-293T cells were transfected using the AAV2-GFP helperless packaging system (Cell BioLabs VK-402). Transfection complexes were prepared using PEIpro® (PolyPlus 115-010) and three plasmid DNAs at a 2:1 ratio (PEIpro:DNA), with a mass ratio of the three plasmid DNAs of 1:1:1 (pAAV-GFP:pHelper:pAAV-RC2). The total amount of plasmid DNA in each container was calculated based on the total surface area, with a predefined DNA density of 0.18 µg / cm². 2 To perform transfection, cells were stored at 22K / cm². 2 Inoculation at 2.5 m 2 Cells were cultured for 3 days in a bioreactor on a Corning® Ascent® PD system. Before transfection, the cell culture medium in the MCV was replaced with an equal volume of fresh complete medium, and the pH was adjusted from 7.2 to 7.0. After the system stabilized for 2 hours, a DNA / PEIpro mixture solution (containing 4.5 mg of total DNA) was injected into the MCV. The DNA complex solution was freshly prepared using the protocol recommended by the supplier (Polyplus, Ilkhich, France). 24 hours after transfection, the cell culture medium in the MCV was replaced with an equal volume of fresh complete medium. The transfected cells were cultured for another 2 days, followed by collection using in situ lysis.

[0159] After culture, cells were collected manually or automatically. For manual collection, three adjacent discs were removed from five different locations in the bioreactor and incubated on a track-guided shaker in a culture dish with 20 mL of aculase for 30 minutes. Finally, the cells were quantified using Vi-Cell and used to estimate the total cell count in the bioreactor. For automated collection, the bioreactor was emptied after culture and washed with DPBS buffer. Then, the cells were flowed in the bioreactor with aculase solution for a specified time. Finally, the cells were washed away, collected, and counted.

[0160] Mathematical Model

[0161] According to the mass balance perfusion bioreactor model, OUR is the DO concentration gradient (C0) as the cell culture medium passes through the bioreactor. 进 -C 出 The function of the culture medium flow rate Q:

[0162] (13)

[0163] Where C 进This refers to the concentration of DO in the culture medium entering the bioreactor (the same as the DO in the MCV, referred to as MCV DO), C 出 Q is the concentration of DO leaving the bioreactor (referred to as FBR DO), and Q is the flow rate (usually expressed in mL / min).

[0164] Assuming the average oxygen consumption rate q per cell o2 Constant and independent of cell cycle and culture conditions. 14 OUR is a direct function of the cell number N:

[0165] (14)

[0166] During cell seeding, the number of cells captured and attached to the mesh surface of the bioreactor continuously increases. Cell adhesion is relatively rapid and typically follows a single-phase association process. Therefore, cell attachment kinetics can be obtained by fitting the OUR curve using Equation 2 to the single-phase association.

[0167] According to cellular energy metabolism models, OUR is related to both the oxygen required for biomass maintenance and the oxygen required for growth:

[0168] (15)

[0169] Where N is the number of cells, M O2 Y is the oxygen consumption coefficient used for maintenance. xo It is the oxygen production consumed for cell growth.

[0170] According to the exponential growth model, the number of cells N at a given time t after cell attachment is complete is:

[0171] (16)

[0172] Where N0 is the number of cells seeded, and k is the cell growth rate.

[0173] According to the logistic growth model, the number N of cells at a given time t after cell attachment is completed is:

[0174] (17)

[0175] Where A is the maximum fold increase, equal to N. max / N0, µ max λ is the maximum cell growth rate, and λ is the lag time.

[0176] Results

[0177] The OUR curves of the HEK293T AAV2 bioproduction experiment and the HEK293 culture experiment were compared below. 2.5 m2 The HEK293T AAV2 bioproduction experiment in the bioreactor consisted of four phases: cell seeding, growth, transfection, and production. The cell seeding phase lasted approximately 3 hours, followed by 3 days of growth, 1 day of transfection, and 2 days of production. Throughout the experiment, the relative dissolved oxygen (DO) in the MCV was maintained at 100% (actual concentration 18.5%). However, the DO in the culture medium leaving the bioreactor initially decreased until it reached approximately 20%, and was then maintained at this level by increasing the flow rate. Figure 14A In comparison, 5 m 2 The HEK293 cell culture experiment in the bioreactor consisted of only two phases: cell seeding and growth. Due to the relatively low oxygen consumption rate of HEK293, the DO (displacement) protocol differed. Here, the relative DO in the MCV was initially maintained at 50% (actual concentration 9.25%) and slowly increased to 100% over time, while the DO in the culture medium leaving the bioreactor initially decreased until reaching approximately 20%, and was then maintained at this level by increasing the flow rate. Figure 14B ).

[0178] Interestingly, OUR continues to increase over time, but exhibits several distinct phases in the HEK293T experiment. Figure 15A This indicates that transfection and AAV2 production do indeed have some impact on nutrient requirements and cell growth. In contrast, OUR also showed two different phases in the HEK293 culture experiment ( Figure 15B This may be due to the increased heterogeneity of dissolved oxygen (DO) throughout the bioreactor as cells grow to high density over time.

[0179] Next, the possibility of using OUR curves to determine cell attachment kinetics was examined. To this end, three different methods were developed and used to fit OUR curves to a typical monophasic association model used for cell adhesion. What these methods have in common is how to estimate the oxygen consumption rate q per cell. o2 However, the difference lies in how the cell number is determined. The first method uses the initial seeded cell number. This method assumes that there is little or no cell division during the cell attachment phase, which is generally acceptable. The second method uses the historical normal growth kinetics of a specific cell line to estimate the cell number over time. The third method fits the OUR to a logistic growth model to estimate the cell number over time.

[0180] The results showed that, for AAV2 production experiments, regardless of the method used to estimate cell number, q o2 Initially, they all increase from a significant negative value and quickly reach a stable state. Figures 16A-16CThe initial negative value was expected because the cells were added directly to the MCV and began consuming the DO in the MCV, resulting in DO concentrations in the MCV slightly below saturation; however, the DO in the culture medium leaving the bioreactor was still close to saturation levels (approximately 100%). After cell attachment was complete, q o2 The observed relatively stable values ​​are consistent with those reported in the literature for cells cultured in stirred tank bioreactors, indicating that the oxygen consumption rate per cell approaches constant over time. The results also show that regardless of the cell number used, q o2 The values ​​fit well using a single-phase association model. Furthermore, the cell attachment kinetic parameters generated by all three methods are comparable to those obtained using offline methods. Figure 16D Using q o2 The ability to determine cell attachment kinetics lies primarily in the different oxygen consumption rates between suspended and adherent cells, and the increase in oxygen uptake in the bioreactor as more cells adhere to the mesh surface.

[0181] exist Figure 17 In this study, cell attachment dynamics are determined by fitting the oxygen consumption rate q. o2 (ac) or obtained by using an offline cell counting protocol (d) in a bioreactor to determine the percentage of cells over time using a single-phase association model. For (ac), q o2 The calculation method is to divide OUR by the number of cells, i.e., the initial seeding number (a), the estimated number of cells determined using the logistic model (b), or the estimated number of cells determined using the historical growth kinetics of HEK293T cells (c).

[0182] In addition, q-based o2 The model demonstrates broad applicability for cell attachment dynamics modeling. Tables 7 and 8 summarize the cell attachment dynamics parameters obtained using the model, compared with those obtained using the model for 2.5 m. 2 HEK293T cells or 5 m in Ascent bioreactor 2 Parameters obtained using a typical offline protocol for HEK293 cells in the Ascent bioreactor were compared. The results showed that for all culture experiments, cell attachment kinetics obtained using the oxygen consumption rate were comparable to those obtained offline, assuming almost no cell growth in the first seven hours after cell seeding.

[0183]

[0184] Table 7. Cell attachment kinetic parameters obtained using the oxygen consumption rate protocol and the offline cell counting protocol used for HEK293T cells.

[0185]

[0186] Table 8. At 5 m 2 Cell attachment kinetic parameters were obtained using an oxygen consumption rate protocol and an offline cell counting protocol for HEK293 cells in the Ascent bioreactor.

[0187] Interestingly, HEK293T cells showed significant stability. o2 The value was consistent across different experiments, but higher than the stable value of HEK293 cells. o2 This is expected, as HEK293T cells grow at a faster rate than HEK293 cells, thus requiring more oxygen.

[0188] Next, both exponential and logistic models were applied to study the cell growth curves of the two cell lines. To test the robustness of model-driven biomass prediction, four different cell culture runs were performed and analyzed for each cell line. Figure 18 This demonstrates an example of a cell culture experiment, in which 5 m 2 The bioreactor was seeded with 1.11E+09 HEK293T cells and cultured for 3 days. The results showed that the predicted cell number using the logistic model described in Equation 16 was 1.27E+10, which was close to the number of cells collected (1.06E+10).

[0189] Tables 9 and 10 summarize the main results of eight culture experiments, with four experiments per cell line. The results confirm that both models provide relatively acceptable accuracy in predicting biomass. It is noteworthy that for some culture experiments, as shown in the two tables, the data fit quality was not very good, resulting in relatively poor predictions. This is because these experiments encountered some issues due to suboptimal pump rates when extremely high cell densities were reached on the last day of culture.

[0190]

[0191] Table 9. Conditions and cell numbers collected for four different cell culture experiments of HEK293T, and cell growth parameters and cell numbers at collection predicted using two different growth models.

[0192]

[0193] Table 10. Conditions and cell numbers collected in four different cell culture experiments using HEK293, and cell growth parameters and cell numbers at collection predicted using two different growth models.

[0194] Next, the feasibility of OUR-based modeling to predict biomass throughout AAV2 production operations was examined. This is important because there is no solution for monitoring and estimating cell growth behavior and biomass throughout the entire bioproduction process. Figure 6 Demonstrated its use at 2.5 m 2 An example of AAV2 bioproduction experiments using HEK293T cells cultured in a bioreactor. This bioproduction run started with 5.67E+08 cells, followed by three days of growth, one day of transfection, and finally two days of AAV production. Results showed that the OUR model fit well with the logistic regression model (…). Figure 17 However, the exponential model did not fit well (data not shown), yielding a final cell number of 1.10E+10, close to the collected cell number (9.63E+9). Table 11 summarizes the results obtained from three different AAV2 bioproduction experiments. The results confirm that the logistic model can estimate the cell number fairly accurately at the end of the bioproduction experiment. In summary, these results indicate that OUR-driven modeling can effectively simulate cell growth curves and predict biomass at the end of culture or bioproduction.

[0195]

[0196] Table 11. Culture parameters and predicted cell numbers for HEK393T cell transfection experiments.

[0197] Finally, OUR-driven modeling is applied to predict biomass in real time using two different methods. These two methods share the commonality of predicting the cell number at a given time, but differ in how q is estimated. o2 The number of cells at a given time is calculated by dividing the corresponding OUR value by q. o2 Values ​​were obtained. The first method used historical normal values ​​obtained for the cultured cell lines. o2 Values. For example, the historical normal average q of HEK293T and HEK293 cells found internally. o2 Value (called q) o2-细胞系 The values ​​were 2.20 ± 0.17E-13 (n = 9) and 1.56 ± 0.23E-13 (n = 4) mol / hr / cell, respectively. For a given cell line cultured in the Ascent bioreactor under optimal and conventional culture conditions, it is reasonable to assume q o2 The value should be within a small range. The advantage of this method is that it can not only predict biomass in real time, but also identify potential abnormal cell growth in culture and bioproduction operations due to nutrient shortages or other operational errors.

[0198] The second method uses stable cells obtained within seven hours after cell seeding to begin the culture experiment.o2 Value (called q) o2-稳定 The advantage of this method is that it is sensitive and specific to cultured cells. It is known that intrinsic biological characteristics of cells (e.g., passage) or environmental factors (e.g., culture medium composition, dissolved oxygen (DO), pH, temperature) can directly influence cell growth patterns; cell growth exhibits some variability, which can lead to culture condition-dependent variations. o2 With this in mind, it is possible to obtain culture-specific q values ​​after culturing cells for a short period of time. o2 For the Ascent bioreactor, it was found that both cell lines typically reached a stable state within approximately five hours after inoculation. o2 Value. Reaching a stable q o2 The timing of the value may be related to the cell line.

[0199] Figure 19 Cell growth curves predicted using these two methods were compared with those obtained by fitting the entire OUR curve using a logistic model. Here, again using 2.5 m... 2 Taking HEK293T cells cultured in the Corning Ascent bioreactor as an example, the same AAV2 bioproduction experiment was conducted. Results showed that both methods produced acceptable real-time prediction accuracy. For logistic modeling, q... o2-细胞系 and q o2-稳定 The predicted cell numbers at the end of culture were 1.10E+10, 1.33E+10, and 1.19E+10, all of which were comparable to the number of cells collected (9.63E+09).

[0200] Biomass is one of the most critical parameters for monitoring bioproduction using mammalian cell culture; the lack of robust biomass monitoring methods is considered a drawback of large-scale adherent cell culture. Recently, the effectiveness of model-driven biomass prediction based on substrate consumption has been established and demonstrated. However, these methods cannot be used to determine cell attachment kinetics due to minimal substrate consumption during cell seeding; furthermore, they cannot be used to predict post-transfection biomass, as transfection may affect substrate consumption behavior. Here, oxygen uptake and consumption are used as indirect indicators of cell attachment and growth kinetics.

[0201] Molecular oxygen is a key regulator of cell proliferation and differentiation and a crucial substrate for cellular metabolism. Oxygen is also one of the most important variables and parameters for monitoring and developing biological processes. For the Ascent bioreactor, several oxygen biosensors were applied to monitor the immediate MCV (molecular volume) leaving the bioreactor and the DO (dissolved oxygen) levels in the culture medium to achieve automated control and process optimization. Different operating protocols were developed and optimized for HEK293T and HEK293 cell cultures. For HEK293T cells, the DO in the MCV was maintained at 18.5% throughout the culture process, while the DO in the culture medium leaving the bioreactor was ultimately targeted to 3.7% by increasing the flow rate. Figure 9 a). In contrast, for HEK293 cells, the DO in the MCV started at 9.25% and gradually increased to 18.5% as the cells continued to grow, while the DO in the culture medium immediately leaving the bioreactor was ultimately targeted at 3.7% by increasing the flow rate. Figure 9 b).

[0202] OUR monitoring has proven to be a direct method for estimating the viable cell density of microorganisms and mammalian cells in suspension cultures. Furthermore, OUR is closely correlated with the physiological state of cells. Due to its non-invasiveness and high frequency of measurement, OUR curves can be adjusted in real time, making it well-suited for designing nutrient feeding strategies. However, the use of OUR to predict the biomass of cell cultures in fixed-bed bioreactors has not yet been explored.

[0203] Here, a mathematical model was developed and its effectiveness in determining the cell attachment and growth kinetics of HEK293T and HEK293 cells cultured in the Ascent bioreactor was demonstrated. Biomass prediction based on oxygen consumption rate in real time and determination of the culture status of mammalian cells were also shown. The OUR-driven modeling developed here not only enables non-invasive, real-time, and automated monitoring but also provides acceptable accuracy in determining cell attachment and growth kinetics and predicting biomass in real time. Furthermore, OUR-driven modeling is compatible with both cell growth and viral vector bioproduction experiments, scalable to different feed programs and operations, and, more importantly, independent of different feed programs and operations (e.g., culture medium replacement, pH adjustment, transfection). This study demonstrates that OUR-driven modeling can serve as a general indicator of metabolic activity and culture behavior in the Ascent bioreactor system.

[0204] * * * * *

[0205] Depending on the desired system, cell culture substrates can be arranged in a variety of configurations within a culture chamber. For example, in one or more embodiments, the system includes one or more base layers whose width extends across the width of a defined cell culture space within the culture chamber. Multiple layers of the base can be stacked to a predetermined height in such a manner. The base layers can be arranged such that a first side and a second side of one or more layers are perpendicular to the overall flow direction of the culture medium flowing through the defined culture space within the culture chamber, or the first side and a second side of one or more layers may be parallel to the overall flow direction. In one or more embodiments, the cell culture substrate includes one or more base layers with a first orientation relative to the overall flow, and one or more other layers with a second orientation different from the first orientation. For example, each layer may have a first side and a second side parallel to or perpendicular to the overall flow direction, or at an angle between the first and second sides.

[0206] In one or more embodiments, the cell culture system includes a plurality of discrete cell culture substrate blocks configured in a filled bed, wherein the length and / or width of each substrate block is relatively small relative to the culture chamber. As used herein, each substrate block is considered to have a small length and / or width relative to the culture chamber when the length and / or width of the substrate block is about 50% or less of the length and / or width of the culture space. Therefore, the cell culture system may include a plurality of substrate blocks filled in a culture space in a desired arrangement. The arrangement of the substrate blocks may be random or semi-random, or may have a predetermined order or arrangement, for example, the substrate blocks may be oriented in a generally similar orientation (e.g., horizontal, vertical, or at an angle between 0° and 90° relative to the overall flow direction).

[0207] As used herein, “defined culture space” refers to the space within a culture chamber occupied by the cell culture medium, in which cell seeding and / or culture will take place. Defined culture space may substantially fill the entire culture chamber or may occupy a portion of the space within the culture chamber. As used herein, “overall flow direction” is defined as the direction of the overall mass flow of fluid or culture medium through or across the cell culture medium during cell culture and / or during the inflow or outflow of culture medium from the culture chamber.

[0208] In some embodiments of this disclosure, only a single overall flow direction exists within the defined culture space, packed bed, and / or bioreactor vessel, such that the liquid or culture medium flow primarily in one direction from the bioreactor inlet through the packed bed to the bioreactor outlet. The liquid or culture medium flow is not interrupted by any complex flow paths within the packed bed space and travels primarily in one direction through the packed bed. This avoids the complex flow paths used in some conventional bioreactors, where flow spacers, baffles, or channels are used to aid in distributing cell culture medium through the cell culture substrate, often due to the inherent non-uniformity of the bioreactor or cell culture substrate. However, in embodiments of this disclosure, these complex flow paths are not necessary, and the culture medium flow can be maintained in a single direction from the bioreactor inlet to the bioreactor outlet. The foregoing is not intended to exclude the use of flow distributor plates at the inlet and outlet of a bioreactor plate, which can be used to distribute fluid across the width of the bioreactor vessel and / or control the pressure differential within the reactor, without additionally affecting the overall flow direction through the cell culture space within the packed bed and / or bioreactor vessel.

[0209] One or more embodiments of a packed bed cell culture substrate may include a substrate material configured to have a uniform and ordered porous structure. The substrate may be referred to as a “structurally determined” substrate, meaning that the substrate has a physical structure that is not random but ordered according to determined parameters. In one or more embodiments, a structurally determined substrate includes a plurality of openings defining the porosity of the substrate, said plurality of openings being arranged in a regular or uniform pattern in each substrate block or layer. In one or more embodiments, a packed bed cell culture substrate may include a woven cell culture mesh substrate, without any other form of cell culture substrate disposed in or dispersed in the cell culture substrate. That is, the woven cell culture mesh substrate of the embodiments of this disclosure is an effective cell culture substrate, eliminating the need for irregular nonwoven substrates of the type used in existing solutions. This simplifies the design and construction of cell culture systems while providing high-density cell culture substrates with other advantages discussed herein regarding flow uniformity, collectability, etc.

[0210] In one or more embodiments, a matrix is ​​provided having a structurally defined surface area for adherent cell attachment and proliferation, good mechanical strength, and forming a highly uniform, multi-layered, interconnected fluidic network structure when assembled in a packed bed or other bioreactor. In certain embodiments, a mechanically stable, non-degradable woven mesh can be used as a substrate to support adherent cell production. The cell culture substrates disclosed herein support adherent-dependent cell attachment and proliferation at high volumetric density. Such substrates achieve uniform cell seeding and efficient collection of cells or other products from the bioreactor. Furthermore, embodiments of this disclosure support cell culture to provide uniform cell distribution during the seeding step and achieve monolayer or multilayer confluence of adherent cells on the disclosed matrix, avoiding the formation of large and / or uncontrollable 3D cell aggregates with restricted nutrient diffusion and increased metabolite concentrations. Thus, the matrix eliminates diffusion limitations during bioreactor operation. Additionally, the matrix enables easy and efficient cell collection from the bioreactor. The structurally defined matrix of one or more embodiments enables complete cell recovery from a packed bed of a bioreactor and continuous cell collection.

[0211] High flow resistance uniformity is achieved throughout the matrix or packed bed by using a culture substrate with a sufficiently rigid structure. According to various embodiments, the substrate can be deployed in single-layer or multi-layer configurations. This flexibility eliminates diffusion limitations and enables uniform delivery of nutrients and oxygen to cells attached to the substrate. Furthermore, the open substrate has no cell-trapping areas in the packed bed configuration, allowing for complete collection of highly viable cells at the end of culture. The substrate also enables uniform filling of the packed bed, enabling direct scaling from process development units to large-scale industrial bioprocessing units. The ability to collect cells directly from the packed bed eliminates the need to resuspend the substrate in agitated or mechanically vibrating containers, which increases complexity and can potentially impose harmful shear stress on the cells. Moreover, the high packing density of the cell culture substrate allows for high bioprocess productivity in manageable volumes at an industrial scale.

[0212] Compared to existing cell culture substrates (i.e., nonwoven substrates formed from randomly ordered fibers) used in cell culture bioreactors, embodiments of this disclosure include cell culture substrates with a defined and ordered structure. This defined and ordered structure allows for consistent and predictable cell culture results. Furthermore, the substrate has an open porous structure that prevents cell trapping and allows cells to flow uniformly through the packed bed. This configuration enables improvements in cell seeding, nutrient delivery, cell growth, and cell collection. According to one or more specific embodiments, the matrix is ​​formed from a substrate material with a thin, sheet-like structure having first and second sides spaced apart by a relatively small thickness, such that the sheet thickness is small relative to the width and / or length of the first and second sides of the substrate. Additionally, a plurality of pores or openings are formed through the thickness of the substrate. The substrate material between the openings has a certain size and geometry, thereby allowing cells to adhere to the surface of the substrate material as if it were an approximately two-dimensional (2D) surface, while also allowing sufficient fluid to flow around the substrate material and through the openings. In some embodiments, the substrate is a polymer-based material and may be shaped as a molded polymer sheet; a polymer sheet with perforations through its thickness; multiple filaments fused into a mesh layer; a 3D-printed substrate; or multiple filaments woven into a mesh layer. The physical structure of the substrate has a high surface area to volume ratio for culturing adherent cells. According to various embodiments, the substrate may be arranged or filled in a bioreactor in some of the ways discussed herein for uniform cell seeding and growth, uniform culture medium perfusion, and efficient cell collection.

[0213] According to one or more embodiments, the cell culture substrate may be one of the cell culture substrates and / or base materials disclosed in the following: U.S. Patent Application Nos. 16 / 781,685; 16 / 781,723; 16 / 781,764; 16 / 781,807; 16 / 781,847; 16 / 781,883; and 16 / 765,722, all of which are incorporated herein by reference in their entirety.

[0214] According to some embodiments, a method for cell culture using a bioreactor with a substrate is also provided for the bioprocessing production of therapeutic proteins, antibodies, viral vaccines, or viral vectors.

[0215] The cell culture substrates and bioreactor systems provided offer numerous advantages. For example, embodiments of this disclosure can support the production of any of a variety of viral vectors, such as AAV (all serotypes) and lentiviruses, and can be used for both in vivo and in vitro gene therapy applications. Uniform cell seeding and distribution maximize viral vector yield per container, and these designs also allow for the collection of live cells, which is highly useful for seed amplification consisting of multiple amplification stages using the same platform. Furthermore, the embodiments described herein can be scaled from process development scale to production scale, ultimately saving development time and costs. The methods and systems disclosed herein also allow for the automation and control of cell culture processes to maximize vector yield and improve reproducibility. Finally, compared to other cell culture solutions, achieving viral vector production scale (e.g., 10 per batch) is significantly easier. 16 Up to 10 18 The number of containers required for an AAV VG can be significantly reduced.

[0216] The embodiments are not limited to the container rotating about a central longitudinal axis. For example, the container can rotate about an axis that is not at the center of the container. In addition, the axis of rotation can be a horizontal axis or a vertical axis.

[0217] definition

[0218] "Totally synthetic" or "completely synthetic" refers to cell cultures composed entirely of synthetic source materials and containing no animal-derived or animal-origin materials, such as the surface of microcarriers or culture containers. The disclosed totally synthetic cell cultures eliminate the risk of foreign contamination.

[0219] The terms “include”, “includes”, or similar terms mean that something is included but not limited to; that is, inclusive rather than exclusive.

[0220] "User" means anyone who uses the systems, methods, articles or kits disclosed herein, and includes anyone who cultures cells to collect cells or cell products, or uses cells or cell products cultured and / or collected in accordance with the examples herein.

[0221] When describing embodiments of this disclosure, the term "about" is used to modify, for example, the amount, concentration, volume, processing temperature, processing time, yield, flow rate, pressure, viscosity, and similar values ​​and ranges thereof of a component in a composition, or the size and similar values ​​and ranges thereof of a component, to refer to variations in numerical quantities that may occur, for example, due to: typical measurement and processing procedures used to prepare materials, compositions, complexes, concentrates, components, articles, or formulations; unintentional errors in these procedures; differences in the manufacture, source, or purity of starting materials or components used to carry out the method; and similar considerations. The term "about" also covers amounts that differ due to aging of compositions or formulations having a particular initial concentration or mixture, and amounts that differ due to mixing or processing of compositions or formulations having a particular initial concentration or mixture.

[0222] The terms “optional” or “optionally” mean that the event or situation described below may or may not occur, and the description includes instances where the event or situation occurs as well as instances where the event or situation does not occur.

[0223] Unless otherwise stated, the indefinite articles “a” or “an” and their corresponding definite articles “the” as used herein mean at least one species, or one or more species.

[0224] Abbreviations well known to those skilled in the art may be used (e.g., “h” or “hrs” for hour, “g” or “gm” for gram, “mL” for milliliter, and “rt” for room temperature, “nm” for nanometer, and similar abbreviations).

[0225] The specific and preferred values ​​and ranges disclosed regarding components, ingredients, additives, dimensions, conditions, and similar aspects are for illustrative purposes only; they do not exclude other defined values ​​or other values ​​within the defined range. The systems, kits, and methods disclosed herein may include any or any combination of the values, specific values, more specific values, and preferred values ​​described herein, including explicit or implicit intermediate values ​​and ranges.

[0226] Unless otherwise expressly stated, it is not intended to interpret any method described herein as requiring its steps to be performed in a particular order. Therefore, no particular order is intended to be implied unless a method claim actually describes the order in which the steps are followed, or unless the claims or specification otherwise specifically state that the steps are limited to a particular order.

[0227] It will be apparent to those skilled in the art that various modifications and variations can be made without departing from the scope or spirit of the disclosed embodiments. Since modifications, combinations, sub-combinations, and variations of the disclosed embodiments incorporated into the spirit and intent of the embodiments will be apparent to those skilled in the art, the disclosed embodiments should be construed as including all contents within the scope of the appended claims and their equivalents.

Claims

1. A method for monitoring biomass during cell culture in a bioreactor, the method comprising: The cells are cultured in the bioreactor using a cell culture medium that has been perfused through the bioreactor; Measure at least one of the cell nutrients and cell byproducts in the cell culture medium; The rate of consumption of the cell nutrients and the rate of accumulation of the cell by-products are determined; The number of cells in the bioreactor at a specified culture time is predicted based on at least one of the consumption rate and the accumulation rate.

2. The method of claim 1, wherein the bioreactor is a fixed-bed bioreactor comprising a substrate configured to culture cells attached to the surface of the substrate.

3. The method according to claim 1 or claim 2, wherein at least one cell nutrient is glucose or glutamine.

4. The method according to any one of claims 1 to 3, wherein at least one cell byproduct is lactate or ammonia.

5. The method according to any one of claims 1 to 4, wherein the cell culture medium is a cell culture medium rich in glucose or glutamine.

6. The method according to any one of claims 1 to 5, wherein measuring at least one of the cell nutrients and the cell by-products in the cell culture medium comprises performing multiple measurements of the cell nutrients or the cell by-products, the interval between the multiple measurements being less than the doubling time of the cells in the cell culture.

7. The method of claim 6, wherein the measurement interval is greater than or equal to a minimum interval time, the minimum interval time being the time during which the change in the level of the cell nutrient or the cell by-product is greater than the measurement tolerance for measuring the cell nutrient or the cell by-product.

8. The method of claim 7, wherein the minimum interval time is greater than or equal to about 30 minutes, 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours, 11 hours, 12 hours, 13 hours, 14 hours, 15 hours, 16 hours, 17 hours, 18 hours, 19 hours or 20 hours.

9. The method according to any one of claims 1 to 8, wherein measuring the cell nutrients in the cell culture medium comprises measuring the cell nutrients multiple times daily during the cell culture.

10. The method according to any one of claims 1 to 9, wherein measuring the cell by-products in the cell culture medium comprises measuring the cell by-products multiple times daily during the cell culture.

11. The method according to any one of claims 1 to 10, wherein predicting the number of cells comprises calculating the biomass at the specified culture time using a mathematical model.

12. The method according to any one of claims 1 to 11, wherein the measurement comprises using an inline sensor in the perfusion line of the cell culture medium.

13. The method according to any one of claims 1 to 11, wherein the measurement comprises an offline measurement using a sample of the cell culture medium.

14. The method according to any one of claims 1 to 13, further comprising, after determining the consumption rate and the accumulation rate, comparing at least one of the consumption rate of the first cellular nutrient and the accumulation rate of the first cellular byproduct with at least one of the consumption rate of the second cellular nutrient and the accumulation rate of the second cellular byproduct.

15. The method of claim 14, wherein the comparison comprises comparing the rate of glucose consumption with the rate of glutamine consumption.

16. The method of claim 14 or claim 15, wherein the comparison comprises comparing the rate of glucose consumption with the rate of ammonia accumulation.

17. The method according to any one of claims 14 to 16, wherein the comparison comprises comparing the rate of lactate accumulation with the rate of glutamine consumption.

18. The method according to any one of claims 14 to 17, wherein the comparison comprises comparing the accumulation rate of lactate with the accumulation rate of ammonia.

19. The method according to any one of claims 14 to 18, further comprising determining abnormalities in the cell culture based on the comparison.

20. The method according to any one of claims 1 to 19, further comprising inoculating cells in the bioreactor at a certain inoculation density.

21. The method according to any one of claims 1 to 20, further comprising supplying fresh cell culture medium to the bioreactor.

22. The method of claim 21, wherein the measurement comprises a first measurement performed at least one hour after the supply of the fresh cell culture medium.

23. The method according to any one of claims 1 to 22, wherein the predicted cell number N t Includes the use of the following equation: N t = N 接种 e kt Where N 接种 The number of cells inoculated into the bioreactor is given by k, the cell growth rate is given by t, and the time for predicting the number of cells is given by t.

24. A method for real-time prediction of cell culture biomass in a bioreactor, the method comprising: The glucose concentration of the cultured cells was measured at predetermined time intervals; At least one mathematical model for biomass prediction is selected based on at least one of the frequency of glucose measurement, availability of maximum achievable cell density, and predetermined lag time of the cell lines cultured in the bioreactor. The measured glucose concentration is fitted according to at least one mathematical model; and Biomass is predicted in real time after the cells have been cultured for a period of time beyond the predetermined lag time of the cultured cell line.

25. A method for monitoring the biomass of cell cultures in a bioreactor, the method comprising: Cells are cultured in the bioreactor using cell culture medium that is perfused through the bed bioreactor; Monitoring the oxygen uptake rate (OUR) of the cells in the bioreactor; and The biomass of the cell culture is predicted based on the oxygen uptake rate.

26. The method of claim 25, wherein the biomass monitoring is performed in real time.

27. The method of claim 25 or claim 26, wherein the bioreactor comprises a fixed-bed bioreactor.

28. The method according to any one of claims 25 to 27, further comprising circulating the cell culture medium using a circulating flow rate.

29. The method of claim 28, further comprising: Provide fixed-bed bioreactors; A culture medium conditioning container is provided, the culture medium conditioning container being configured to contain and condition the cell culture medium; Provides a flow circulation path for fluid connection between the culture medium conditioning vessel and the fixed-bed bioreactor; and The cell culture medium is circulated from the culture medium conditioning container to the fixed-bed bioreactor and back to the culture medium conditioning container using the circulation flow rate.

30. The method of claim 29, further comprising: As the cell culture medium flows from the culture medium conditioning container to the fixed-bed bioreactor, a first dissolved oxygen level in the cell culture medium in the culture medium conditioning container or between the culture medium conditioning container and the fixed-bed bioreactor is measured; and The second dissolved oxygen level in the cell culture medium was measured after the cells left the fixed-bed bioreactor.

31. The method of claim 30, wherein the measurement of the second dissolved oxygen level occurs before the cell culture medium leaves the culture medium conditioning container and proceeds to the fixed-bed bioreactor.

32. The method of claim 30 or claim 31, further comprising providing an initial dissolved oxygen sensor within the culture medium conditioning container.

33. The method according to any one of claims 30 to 32, further comprising providing an outlet dissolved oxygen sensor disposed in the flow circulation path after the cells in the bioreactor and before the culture medium conditioning vessel.

34. The method according to any one of claims 30 to 33, further comprising calculating the oxygen uptake rate (OUR).

35. The method of claim 34, wherein the OUR is calculated according to the following equation: OUR = (C 进 – C 出 )*Q, Where C 进 It is the dissolved oxygen concentration of the cell culture entering the bioreactor, and C 出 Q is the dissolved oxygen concentration leaving the bioreactor, and Q is the circulation flow rate (mL / min).

Images