Skip to main content
NCBI home page
Cytotechnology logoLink to Cytotechnology
. 2022 Oct 12;74(6):669–680. doi: 10.1007/s10616-022-00554-y

Comparative study of commercial media to improve GMP manufacturing of recombinant human interferon β-1a by CHO cells in perfusion bioreactor

Hossein Sedighikamal 1,2,✉, Reza Karimi Mostofi 1,3, Alireza Sattarzadeh 1, Mansour Shahbazi 1, Hossein Aghazadeh 1
PMCID: PMC9652187  PMID: 36389287

Abstract

Chinese hamster ovary cells are the main cellular factories for production of a wide range of recombinant proteins in biopharmaceutical industry. Recombinant human Interferon beta-1a (rh-IFN β-1a), as a cytokine is broadly used to treat multiple sclerosis. In this work, the cell line producing rh-IFN β-1a was studied to improve cell density along with the specific expression. For this reason different cell culture experiments were done using different commercial serum-free media to find the appropriate media providing higher cell density. It was shown DMEMF12, DMEM:ProCHO5, and CHO-S-SFM II led to higher cell density and shorter doubling time. Next, using these media, fed-batch, and perfusion culture with temperature shift were implemented to investigate the best condition for industrial-scale manufacturing of rh-IFN β-1a in terms of higher cell density and product expression yield. The results demonstrated that CHO-S-SFM II media and a thermally biphasic condition provide enhanced expression of rh-IFN β-1a in perfusion bioreactor.

Keywords: Chinese hamster ovary cells, Specific productivity, Rh-IFN β-1a, Fed-batch, Perfusion

Introduction

There is a rapid increase in demand for the manufacture of biopharmaceuticals for human diseases treatment (Ozturk and Hu 2005; Gessani, et al. 2014; Mohsenzadegan et al. 2015; Bandaranayake and Almo 2014; Yusufi et al. 2017). Most of them are complex glycoproteins manufactured in genetically engineered mammalian cells in culture (Li et al. 2010). Interferon beta (IFN-β) is a cytokine having antiviral, anti-proliferative and immune regulatory properties (Li et al. 2010; Palomares et al. 2004) and is used therapeutically by decelerating the progress of disability to multiple sclerosis (MS) (Pollock et al. 2010; Chotteau 2009; Whitford 2006; Puente-Massaguer et al. 2019), possibly by its anti-inflammatory properties and its potential to improve the integrity of the blood–brain barrier (Altamirano et al. 2004; Reuveny et al. 1986). Recombinant IFN-β was originally produced from Escherichia coli in a non-glycosylated form (Li et al. 2010). Later a Methionine-1 deletion and a cysteine-17 to Serine mutation were introduced to enhance the protein stability (Yusufi et al. 2017; Palomares et al. 2004) and this is one of the versions of IFN-β used therapeutically as IFN-β-1b. Recombinant human IFN-β-1a, produced in mammalian cells such as Chinese Hamster Ovary (CHO), is a single glycosylated polypeptide chain that is made by gathering 166 amino acids with a molecular mass of 18.5 kDa and is reached to 22.5 kDa with an N-linked binding of glycan agent to Asparagine 80. This protein has three cysteines (Cys) with one intramolecular disulfide bridge (Cys31–Cys141) and one free cysteine residue (Cys17). The correct formation of the disulfide bond is vitally essential for the biological activity (Gessani, et al. 2014). This protein consists of 40% hydrophobic amino acids so it is extremely hydrophobic in aqueous solutions. Its isoelectric point is slightly basic (7.8–8.9). The amino acid sequence shows four histidine residues in positions 93, 97, 121 and 131, which explains a good affinity to Me+2 ligands (Gessani, et al. 2014).

Chinese hamster ovary (CHO) cells have emerged as a robust factory in bio-synthesis of biologics. Their unique combination of beneficial quality characteristics makes them an ideal cell line to express different types of recombinant proteins. Amongst other useful factors, the fact that they are very easy to grow in large scale cell culture under defined process conditions and allowing human bio-similar post-translational modifications render CHO cells a powerful bioprocess factory. These cells grow well in both suspension and adherent culture, moreover, their tolerance to variations in pH, Oxygen levels, temperature or pressure make them the ideal cell for large-scale manufacturing under good manufacturing (GMP) regulatory (Li et al. 2010).

Problems facing in adherent culture including the need for serum sources which result in viruses and prions risk of contamination, batch to batch variations in product characteristics, and the need for bigger downstream capacity to compensate the added serums as impurity, have made the emergence of suspension culture mode facilitating a predefined and constant critical quality attributes (Chotteau 2009; Whitford 2006; Puente-Massaguer et al. 2019). Moreover, serum-free media with chemically defined components provides a high-throughput media for high performance in fed-batch or perfusion cultures (Altamirano et al. 2004). Depending on the cell density and cell specific expression, the culture mode can diversify into batch, fed-batch, and perfusion mode. While the batch mode is easy to operate, reliable, less risk of contamination, it rarely can last more than 10 days as no media or feed are added during the culture resulting in nutrients depletion or/and toxic metabolites accumulation, which is not desired for the cells or the products. Fed-batch cultivation of mammalian cells is typically applied for mass production of biologicals as a simple process to increase cell density and to extend the culture lifetime for a higher product concentration. In this method the operation is as simple as batch mode and flexible to be implemented in facilities without main equipment modifications. However, its output in rising the product titer is dependent on the efficiency of the feeding strategies to prevent nutrients depletion and toxic byproduct’s accumulation. To enhance the cell growth or lessen the cell death for high product expression, the critical nutrition and biochemical parameters including osmolality and carbon dioxide should be kept within suitable levels (Ozturk and Hu 2005). Rational adding of feed or media results in higher product yields, culture longevity, higher viability, upper product concentration, and reduction of toxic metabolites including lactate and ammonium.

While batch and fed-batch processes are the most commonly used production systems for mammalian cells, nowadays, the perfusion processes are being increased widely (Karst et al. 2018). In the perfusion mode, the media is added and the supernatant including the product is harvested cell-freely and constantly during long culture days (Kumar et al. 2007). The perfusion bioreactors are used in culture of mammalian cells due to several advantages: it allows prolonging healthy cultures, potentially at high cell density, as well as a short residence time of product in bioreactor. This is more favorable for the product quality and is mandatory for the production of unstable or highly hydrophobic glycoproteins (e.g., IFN-β, factor VIII) (Ozturk and Hu 2005). The continuous addition of fresh media and spent medium removal provide an easy method to decrease accumulated by-products, like ammonia and lactate, which negatively affect the cell growth (van Mastrigt et al. 2018). Furthermore, the cell retaining devices used in perfusion system such as spin filter and alternating tangential filtration remove the necessity for the cell clarification at the end of process. It is more efficient to keep the viable and productive cells with slow growth inside the bioreactor rather than growing the cells in new batch. Another advantage of the perfusion mode is the use of smaller bioreactor compared with fed-batch processes (Ozturk and Hu 2005; Zoro and Tait 2017).

In perfusion process, Ozturk and Hu (2005) introduced an applicable term called cell specific perfusion rates (CSPR). The CSPR is the volume of medium renewed per cell and day. The principle was that if a CSPR providing sufficient cell environment was known for a moderate cell density, the same CSPR could be used for very high cell densities (Ozturk and Hu 2005).

The choice of cell culture media is extremely important, affecting the success of cell culture experiments (Arora 2013; Yang 1991). The selection of the media basically relies on the type of cells to be cultured and also the protein of interest to be produced (Shani et al. 2021; Yang and Xiong 2012). Different cell types have highly specific growth requirements, so, the most appropriate media for each cell type must be optimized experimentally (Bedoya-López et al. 2016). The complexity of the compositions of cell culture media provides many challenges to optimize individual components of media. On the other hand, to enhance the gene of interest expression and controlled cell growth the cultivation temperature can be reduced either suddenly or gradually, called cold shock or hypothermic conditions. In this way high-proliferation state of the cell switches to high-production state (Tharmalingam et al. 2008). The hypothermic conditions could also cause longevity of culturing, reducing the consumption rate of nutrients, releasing fewer apoptotic enzymes, improving protein bioactivity, and avoid hydrophobic product aggregation (Rodriguez et al. 2008). This strategy has been widely applied for a broad range of recombinant proteins, e.g., human Interferon beta, alkaline phosphatase, Erythropoietin, and Fab antibodies (Rodriguez et al. 2008, 2010; Spearman et al. 2007; Masterton and Smales 2014). The mechanisms of how temperature affects the cells are not well established yet (Pereira et al. 2018), but with no doubt, the temperature reduction during the cultivation changes a wide range of mechanisms in cell metabolism (Roobol et al. 2011). This shock keeps the cells in the G1 phase. This arresting happens due to the activation of the ataxia-telangiectasia mutated and Rad3-related kinase (ATR)–p53–p21 pathway resulting in the activation of the p53 signaling pathway (Ohnishi et al. 1998; Yoon et al. 2003). Three temperature set points were investigated in high-density perfusion cultivation of CHO cells (van Mastrigt et al. 2018; Chuppa et al. 1997). They studied CHO cells under 34 °C, 35.5 °C, and 37 °C temperature. The results showed that reducing the temperature to 34 °C lead to higher cell density in oxygen-limited bioreactors, a lower perfusion rate, improved product quality, and easier pH control (Kumar et al. 2008; Xu et al. 2019). Conversely, in another research done by Xu et al (2019) cultivating the Hybridoma cells at 33 °C lessens the productivity of monoclonal antibodies (Lin et al. 2017).

During the last years, different serum-free and chemically defined CHO-specific media have become introduced optimized by different feeds. However, there is an urgent need to find their effects on cell metabolism and protein quality when using different combinations of commercial basal media and feeds. In this study different commercial serum-free media were selected in the logarithmic phase of cell growth curve providing higher viable cell count (VCC) and longer culture duration for industrial manufacturing of rh-IFN. Batch, fed-batch and perfusion culture modes and combinations of batch and perfusion modes with temperature shift, i.e., thermally biphasic conditions (37–32 °C) are implemented to achieve higher VCC, and expression. It is worth mentioning that the entire experiment was conducted triplicate and all the results shown in the Figures and Tables are means ± standard deviation (SD) from the three trials.

Materials and methods

Cell line and media

A CHO-K1 cell line (Cellca GmbH, Laupheim, Germany), originally from organism of Cricetulus griseus, with epithelial morphology, and deficient in dihydrofulate reductase (DHFR−) producing recombinant human Interferon beta 1a were used in this study. The cells stored in liquid Nitrogen tank with temperature lower than − 150 °C were initially thawed in spinner flask (Corning, US) and sub-cultured in different commercial serum free media including Power CHO 2®, ProCHO5®, DMEM/F12, RPMI 1640, 50:50 RPMI:DMEM/F12, 70:30 DMEM/F12: ProCHO5, EX-CELL® CD CHO, and CHO-S-SFM II®.

Batch cultures

In all batch mode experiments the cultures were inoculated at initial VCC of 0.5 × 106 cells/mL, with a working volume of 100 mL. The cells were incubated in a humidified incubator (RH > 80%), with 5% CO2 at 37 °C. During the cultures, ~ 0.5 mL sample were taken aseptically every day for cells counting and measure of glucose, lactate, and ammonia concentrations. VCC, total cell count (TCC), and viability were measured using the Trypan blue exclusion method via a hemocytometer and phase-contrast microscopy. Glucose, lactate, and ammonia concentrations were quantified based on commercial kits (Biorexfars, Iran). In all cases, the cultures were harvested once viability dropped below 70%.

Fed-batch cultures

Fed-batch experiments were grown in 250 mL spinner flasks with initial volume of 100 mL. The inoculum cell density was 0.5 × 106 cells/mL applying the same culture conditions as described for batch cultures. The experiments were supplemented with feeds, based on to the manufacturer’s recommendation to investigate their effect on culture performance. In all experiments, glucose, lactate, and ammonia concentrations measured daily. Similar to batch cultures, all cultures were terminated once the viability dropped below 70%.

Perfusion culture

The perfusion culture was done using cell retention device (spin filter, 10 µm, surface area 800 cm2) with the initial cell density of 0.5 × 106 cells/mL in 10L stirred tank bioreactor (Biozeen, India). Culture temperature was controlled at 37° ± 0.5 °C and cells obtain their oxygen in free and non-compound forms, called dissolved oxygen (DO), a key substrate for growth, production, and maintenance activities. In this work the DO was controlled at 50 ± 5% by introducing sterile air through a microsparger inside the bioreactor. The pH set point was at 7.1 ± 0.1, using the 8% NaHCO3 (Sigma), and CO2 gas through the microsparger. The agitation was done by a marine-blade impellers (diameter = 10 cm) at 120 ± 20 RPM cascaded by DO. The CSPR = 0.10 nL/cell/day were used in all experiments once VCC > 3 × 106 cells/mL. To provide the hypothermic shock (32 °C) the culture temperature was decreased gradually with the rate of 1 °C/6 h, once VCC reached to 10 ± 0.5 × 106 cell/mL.

Protein assay

The concentration of interferon β-1a was measured using commercial ELISA kit (Cygnus Technologies kit). Briefly, 100 µL of anti-CHO:HRP was pipetted in all wells. Then 50 µL of standards, controls and samples were pipetted into wells. Then the wells were covered and incubated on orbital shaker at 400–600 RPM for 2 h at room temperature (20 ± 2 °C). Then the wells were dumped and filled with diluted wash solution by pipetting in ~ 350 µL (repeated 4 times). Any liquid was wiped off from the bottom out side of microliter wells then 100 µL of TMB substrate was pipetted. Afterward, it was incubated at room temperature for 30 min without shaking. At the end, the absorbance of micro plate was read at 450/650 nm and the concentration of IFNβ-1a was calculated based on the standard curve.

Protein bioactivity

The potency of interferon β-1a protein is estimated by comparison of the protein ability to protect A549 cells from a viral cytopathic effect and the same ability of the referenced standard of human IFN β-1a calibrated in International Units (IU). In this way, the sub-cultured A549 cells were exposed to both standard and product, separately, in 96-well plates. The plates were then incubated at 37 °C and 5% CO2 for 20 h. On the next day the encephalomayocarditis virus was added and incubated for 22 h. In the next step the cells were crystal-violet dyed for 1 h at room temperature and then were washed several times. Finally, the sample’s absorbance was read at 570 nm and quantified compared to standard served as a bioassay calibrate with 40,000 IU for measuring the potency of IFN β-1a. Based on pharmacopeia the estimated potency must be within 80–125% of referenced standard.

RNA quantification

Total RNA was isolated from 1 × 106 cells using TRIzol substance (Invitrogen, US) and following the manufacturer’s procedure. The concentration and quality of the total RNA was valued from the absorbance at 260 and 280 nm, measured via a RNA/DNA Calculator (AmerBiosciences, US). The RNA concentration was took from the absorbance at 260 nm, and the RNA quality could be measured from the A260/A280 ratio. A ratio of ~ 1.8 or upper is considered a sign of a high-quality RNA. First-strand cDNA was synthesized from 1 to 5 µg of total RNA by means of Superscript reverse transcriptase (Invitrogen) and the company’s procedure. A 121-bp region of the rh-IFN-β cDNA was amplified by primers F378 (5-GAATGTCCAACGCAAAGCAA) and R498 (5-CTGGGATGCTCTTCGACCTC) by the following real-time PCR analysis. The reaction was done in 50 µl of a buffer having SYBR® Green (Applied Biosystems, US), 0.2 mM dNTPs mix, 2 mM MgCl2, 0.2 µM of each primer, 1 unit of Dynazyme II thermostable DNA polymerase (2 Units/µL; Finnzymes Oy, Espoo, Finland) and 1.5 µL of a 1/10 dilution of the cDNA sample in water. Real-time quantification was completed using the Rotor-Gene 2000 system (Corbett Research, Australia). Cycling settings were 3 min at 95 °C and formerly 40 cycles having 30 s at 95 °C, 30 s at 61 °C and 30 s at 72 °C. The PCR product was established as a single amplicon of the accurate size by melting-curve method and gel electrophoresis. A standard curve was made by running real-time PCR on serial dilutions of IFN-β cDNA standards with known concentration. A graph of Ct against the logarithm of concentration is linear and can be interpolated to catch the concentration of unknown samples. Both samples and standards were run in triplicate and concluding Ct values were considered as the average of the three runs. A 163-bp section of the β-actin cDNA was amplified by primers F3-ACTIN (5-AGCTGAGAGGGAAATTGTGCG) and R9-ACTIN (5-GCAACGGAACCGCTCATT) and the same procedure as defined above for IFN-β, without that the annealing temperature was 59 °C instead of 61 °C.

mRNA quantification

Real-time reverse-transcription (RT) PCR is a consistent technique for precisely quantifying mRNA levels (Bustin 2000). Real-time RT-PCR analyses for quantifying IFN- β and β-actin mRNA were established and used to measure the mRNA existing in CHO cells producing IFN-β. The β-actin gene is frequently used as an internal control once normalizing mRNA levels (Bustin 2000). Furthermore, for CHO cells cultured at temperatures 30–37 °C, β-actin expression has been revealed to remain constant (Furukawa and Ohsuye 1998), and so this housekeeping gene is mainly suitable for the experimental conditions in this work. Additionally, endogenous gene mRNA levels, such as those of glyceraldehyde-3-phosphate dehydrogenase, mitochondrial F1-ATPase subunits 6 and 8, Cyclophilin, and Calmodulin, have been revealed to remain almost constant once shifting from 37 °C to hypothermal conditions, signifying that endogenous gene levels are controlled once temperature is decreased (Frerichs 1998).

Results and discussion

Selection of optimal media in batch culture

Figure 1 shows the growth curves of all commercial media cultured in batch mode, after 10 days. The maximum VCCs are higher in CHO-S-SFM II, DMEM/F12:ProCHO5, and DMEM/F12 in comparison to the other media. Apart from EX-CELL CD and RPMI 1640 media, in all media the cells experienced only 1 day in lag phase. The cells in both CHO-S-SFM II and DMEM/F12:ProCHO5 media were doubled in almost 24 h, while in other media, the doubling time took longer time. The cell viability in CHO-S-SFM II, DMEM:ProCHO5, and DMEMF12 are higher than other, at the end of batch (see Table 1). Additionally, lactate and ammonia concentrations were measured at the end of batch to study their toxic effect on cells growth during the batch. The effects of lactate concentration on cell growth are observed for concentrations of 40 mM or higher as reported by Ozturk and Hu (2005) and the effects of ammonium on cell metabolism are observed from much lower concentration levels, usually 2–4 mM, which are easier to reach in CHO cell culture. In all cases it was observed that lactate and ammonia concentrations were far from their inhibitory concentration.

Fig. 1.

Fig. 1

Open in a new tab

The growth curves of different commercial media cultured in batch mode

Table 1.

The growth parameters of different commercial media in batch mode cell culture. Values are given as means ± SD

Growth parameters SFMs
ProCHO5 DMEM/F12 DMEM/F12:ProCHO5 CHO-S-SFM RPMI1640 RPMI:DMEM/F12 PowerCHO-2 Ex-Cell CD CHO
Viability (%) 84 ± 1 90 ± 1 93 ± 1 94 ± 1 62 ± 1 45 ± 1 79 ± 1 87 ± 1
VCCmax × 106 (cell/ml) 2 ± 0.1 3.8 ± 0.1 3.7 ± 0.2 3.8 ± 0.1 1 ± 0.2 1.5 ± 0.2 1.7 ± 0.3 1.4 ± 0.1
Lag phase (day) 1 ± 0.07 1 ± 0.04 1 ± 0.08 1 ± 0.05 2 ± 0.09 1 ± 0.05 1 ± 0.07 2 ± 0.08
Doubling time (h) 36 ± 4 36 ± 3 24 ± 2 24 ± 1 48 ± 4 48 ± 1 36 ± 2 48 ± 4
Lactate (mM) 25 ± 2 28 ± 2 40 ± 1 18 ± 1 2 ± 0 13 ± 1 21 ± 4 35 ± 3
Ammonia (mM) 3.5 ± 0.2 1.80 ± 0.1 1.48 ± 0.3 2 ± 0.1 1 ± 0.2 1.2 ± 0.4 1.4 ± 0.1 2.8 ± 0.3
Open in a new tab

Fed batch culture

In the next series of experiments, three culture media with the highest amount of VCC were selected as the basal medium and fed-batch experiments were designed accordingly. The method was that media addition (as a feed) was started approximately 48 h before the beginning of the descending trend of the VCC (~ days 3–4). The feed used was the same basal medium and was added to 10% v/v of the cultivation volume. Figure 2 presents the results of fed batch cultivation. As can be seen, by adding only 10% v/v of the feed, the maximum VCC almost doubled and the culture duration was longer than the batch mode.

Fig. 2.

Fig. 2

Open in a new tab

The growth curves of different commercial media cultured in fed-batch mode

IFN β-1a titer enhancement

After finding the medium(s) providing the higher cell density in the logarithmic phase, in this step, in the stationary phase different strategies were implemented to reach higher protein expression. Figure 3 shows these strategies in fed-batch mode. As can be seen, for DMEM/F12 medium the temperature shift and the feeding almost had almost no effect on the IFN-β expression (the mean concentration is ~ 3.5 ± 0.12 µg/mL), while feeding with DMEM/F12:ProCHO5 and CHO-S-SFM II media could increase the IFN-β titer meaningfully (the concentration was up to ~ 7.5 ± 0.38 µg/mL in DMEMF12:ProCHO5 and concentration was up to ~ 9.3 ± 0.17 µg/mL in CHO-S-SFM II). In CHO-S-SFM II medium the temperature shift led to expression improvement over 2.5 folds compared to DMEMF12 medium. The titer increased more when the cells were fed with 1× concentrated media rather than 2× and 4× concentrated media. The same behavior was observed in CHO-S-SFM II medium. The highest protein expression was obtained under biphasic condition (the temperature shift from 37 to 32 °C) and adding a 1× concentrated media as a feed.

Fig. 3.

Fig. 3

Open in a new tab

The effect of different strategies including temperature shift and feed addition on volumetric expression in fed-batch culture mode. On the day 3, the feeding was done with 10% v/v of the initial medium with 1×, 2×, and 4× concentrated medium. Experiments performed at temperature 37 °C was represented by 37 and experiment with temperature shift from 37 to 32 °C was represented by 32. The cell culture temperature was decreased gradually from 37 to 32 °C with the rate of 1 °C/6 h. The sample was taken from each culture once viability is < 75%

Perfusion process

Figure 4 shows the cell growth curves in three selected media. It should be noted that in all three cultures, the CSPR = 0.1 was selected. As can be seen in this figure, cell growth in all three media continues until the 8th day of culture with a similar trend, so at this stage of cell growth it is possible to use any of these three media. On the 8th day, when the VCC exceeded 10 × 106/mL, a temperature shift from 37 to 32 °C was applied within 24 h. As can be seen in the figure, the hypothermal shock caused cell growth to stop for 48 h in all three culture media, however then the cell growth in each medium showed different behaviors. For the DMEMF12:ProCHO5 medium, the cell remains in a static phase for about 1 week without a significant increase in cell density, after which the density of viable cells was decreased so that the batch is finished before day 28. Unlike DMEMF12:ProCHO5 culture medium, in DMEMF12 and CHO-S SFM II, the cells started growing 2 days after the temperature shock and continued upwards until the 18th day of culture. Although for DMEMF12 medium, this trend changed in the 18th day onwards and the cell growth declined afterwards. For CHO-S SFM II, on 18th day onwards, the cell density drop was much slower, so that until the last days, i.e. days > 25, the cell density was still observed above 13 × 106/mL. According to what has been observed, the most suitable culture medium from cell growth point of view is CHO-S SFM II, although its final choice depends on the quantity (product expression) and quality (mainly potency) of the product, which is studied further.

Fig. 4.

Fig. 4

Open in a new tab

The cell growth curves in perfusion bioreactor in three selected media

Figure 5 shows the specific expression versus culture days, in perfusion process. During days 1–8, the cell specific expression in CHO-S SFM II medium was almost 2 times higher compared to other two media at 37 °C, although the cell density these days was not significantly different for these three media. From the 8th day, the temperature shift from 37 to 32 °C increased the cell specific expression in all three media. It has previously been stated (Gessani et al. 2014; Yusufi et al. 2017; Puente-Massaguer et al. 2019; Altamirano et al. 2004) that a decrease in temperature throughout the production period results in higher yields of some proteins and it was confirmed here for rh-IFNβ-1a as well. According to Fig. 5, hypothermic condition increased the qp between 3 and 4 times in all three studied media; so that in CHO-S SFM II medium from 0.1 to 0.15 before the temperature shock reached to 0.3–0.4 on the 12th day. The important point in comparing DMEMF12 and DMEMF12:ProCHO5 media is that while at 37 °C the specific expression of cells in DMEMF12 was higher than DMEMF12:ProCHO5, this trend changed with applying hypothermal conditions and the cell specific expression in DMEMF12:ProCHO5 increased. According to what has been described, the most suitable culture medium in terms of the quantity of IFN produced is CHO-S SFM II medium by applying thermally bi-phasic condition (temperature shift from 37 to 32 °C).

Fig. 5.

Fig. 5

Open in a new tab

The effect of different media on the concentration of rhIFNβ-1a, under temperature shift in perfusion mode

One possible explanation for the temperature effects is that IFN mRNA is more stable at hypothermal temperatures as stated by different groups (Fox et al. 2005). To decide whether the major differences in cell productivity at 32 °C compared to 37 °C might be attributed to the cells having more total RNA at low temperature, this property during the perfusion culture was measured and was found to be constant, irrespective of culture temperature (15.6 ± 3.2 μg/106 cells at 32 °C and 15.6 ± 3.1 μg/106 cells at 37 °C). Therefore the variations in full RNA concentration/cell do not look to be an issue in explaining the increase in specific productivity at hypothermic condition. In next step, the level of IFN- β mRNA (normalized to β-actin mRNA) was also quantified. The analysis was performed at different days before and after temperature shift (days 5 and 11). Figure 6 displays almost one–one correlation between an increase in mRNA levels and in IFN- β specific productivity. The half-life time (t1/2) for IFN- β and β-actin mRNA were quantify at both 32 and 37 °C by watching the mRNA levels at different time period following actinomycin D (a transcription inhibitor) treatment. Without transcription and cell growth, the variation of mRNA level with time is as a result of degradation. Degradation of mRNA is mostly assumed to be first-order respecting mRNA concentration (Leclerc et al. 2002), thus the t1/2, can be found from profile of mRNA level versus time (Ross 1995). The actinomycin D method is well established by Ross (1995), and has been revised later for real-time PCR (Leclerc et al. 2002). In this item the mRNA levels are normalized by the total RNA concentration. The logic for this normalization is that total RNA is made of rRNA and tRNA, which are more stable than mRNA, and their levels will not drop fast after transcription inhibition (Leclerc et al. 2002). On days 5 and 11 of the perfusion culture, actinomycin D was added to several sample of cultures to inhibit transcription. Cell samples were taken 0, 2, 6, 10, 18, 24 and 30 h after actinomycin D adding for quantifying IFN-β and β-actin mRNA values by means of the real-time PCR protocol mentioned above. The concentration of IFN- β and β-actin mRNA were measured against the time, and the corresponding t1/2 values are shown in Table 2. As can be found hypothermic condition has a notable effect on stabilizing mRNA, with the half-lives of both mRNAs increasing by 7–8-times. The β-actin half-life at 37 °C (10 h) is of the same magnitude as the values found for this gene in other cell lines (6–14 h) as mentioned by Leclerc et al. (2002).

Fig. 6.

Fig. 6

Open in a new tab

Specific expression versus with the level of IFN-β mRNA. The specific expression was measured during days 5 and 11 (before and after temperature shift) of the perfusion culture and the IFN- β mRNA levels during the same days were determined and normalized to β-actin mRNA. Samples were measured from two conditions (37 °C and temperature shift from 37 to 32 °C). The ratio of productivity and IFN- β mRNA at 32 °C to their counterparts at 37 °C (labelled ‘37–32 °C) was calculated

Table 2.

mRNA half-lives in perfusion culture mode

Temperature (°C) t1/2 (h)
IFN-β mRNA β-Actin mRNA
37 11.5 ± 1 11 ± 1
32 90 ± 15 74 ± 18
Open in a new tab

Values are given as means ± SD. Half-life (t1/2) values were calculated from the equation: t1/2 = Ln 2/((Ln mRNA1/mRNA2)/(t2 − t1)), assuming first-order degradation rate of mRNA. mRNA1 and mRNA2 are the mRNA concentrations at t1 and t2 times

Figure 7 shows the specific activity for all three culture media, measured once every 4 days. According to what is observed, the specific activity of rh-IFN β-1a protein in CHO-S SFM II showed the highest value compared to other media and are not the same for all three media (the average values are 236, 177, and 193 MIU/mg for CHO-S SFM II, DMEMF12, and DMEMF12:ProCHO5, respectively). In other words, the specific activity can be dependent on the culture medium used; and it can be understood that the components of the culture medium can affect the formation, stability, and activity of interferon protein in terms of primary, secondary, tertiary structures and post translational modifications. Also, according to what is observed in this figure, for all culture media, the temperature shift did not lead to a significant change in the specific activity of interferon.

Fig. 7.

Fig. 7

Open in a new tab

the specific activity for all three culture media, in perfusion mode

Metabolites analysis

Figure 8 shows the glucose and lactate concentration in the culture media during different culture days. As can be seen, with increasing cell density for all three culture media, glucose concentration has decreased from the 1st day, which is in the range of 4–5 g/L, and has decreased to about 1 g/L. By imposing a temperature shock in DMEMF12 and DMEMF12:ProCHO5 media from the eighth to the 16th days of culture the glucose concentration remained almost constant and then in the 19th day onwards, the glucose concentration increased, in other words, the cell consumption decreased and this could be due to Hypothermal conditions and slowing of cellular metabolism associated with cells proliferation. Compared to DMEMF12 and DMEMF12:ProCHO5, CHO-S SFM II medium is slightly different, so that the trend of glucose concentration remained constant only from the eighth to the 10th days, and then with keeping on culture, the concentration of glucose in the medium increased. Figure 8 also shows the variation of lactate metabolite on different days of culture. The trend of changes in this curve is upward for all three culture media until 8–10 days and lactate concentration is constantly increasing, but with hypothermic conditions decreased gently.

Fig. 8.

Fig. 8

Open in a new tab

the glucose and lactate concentration in the culture media during different culture days

In order to study the effect of these metabolites in more detail, the cell-specific rate of consumption and production of glucose (qglc,) and lactate (qlac) are shown in Fig. 9a as a function of culture days in CHO-S SFM II medium. What can be inferred from Fig. 9a is that after the first transient days, the qglc, qlac at 37 °C are significantly higher than the following days at 32 °C, which indicates a high metabolisms associated with high cell growth rate. The qlac decreased very little from 10 days onwards and was almost constant. This trend is slightly different for glucose. From days 3 to 8, the rate of glucose consumption in the cell is constantly decreasing, but from about 18 days onwards, the rate of glucose consumption increased again for a short time and then continued its downward trend again. Also, as Fig. 9b shows, regardless of the early days of the culture, the ratio of lactate to glucose is always less than 1. This value has a constant value on the day of temperature shock until the 11th day. And from that day on, despite some fluctuations until the last days, it has taken an increasing trend. These results fall within the range of formerly published CHO hypothermic culture conditions records. For example, one group found a 46% drop in glucose consumption at 32 °C compared with 37 °C (Yoon et al. 2003). Though another group found a 43% rise in glucose consumption at 32 °C compared with 37 °C (Fox et al. 2003). These results are consistent with those achieved by Fox et al. (2005), who found that specific glucose consumption for their CHO cell line increased by 46% at 30 °C compared with that at 37 °C. Though, the findings are not consistent with those found by Furukawa and Ohsuye (1998), who found that glucose consumption decreased by 24% at 32 °C compared with 37 °C. Thus there are no clear trends regarding the effect of low temperature on CHO nutrient consumption rates. In some cases, the rates are observed to increase at hypothermic condition, while in other cases they decrease. Obviously, more study is essential to comprehend the underlying mechanism for the large rate variances observed in the references.

Fig. 9.

Fig. 9

Open in a new tab

a, b The cell-specific rate of consumption and production of glucose and lactate metabolites as a function of culture days, in CHO-S SFM II medium

Conclusion

The CHO cell line producing interferon beta was cultured in different serum-free media including Power CHO 2®, ProCHO5®, DMEM/F12, RPMI 1640, 50:50 RPMI:DMEM/F12, 70:30 DMEM/F12: ProCHO5, EX-CELL® CD CHO, and CHO-S-SFM II® to provide higher viable cell count (VCC) and thus higher yield of production. For this reason, the cell was cultured in fed-batch and perfusion modes. The results showed that the CHO-S-SFM II, DMEMF12, and 70:30 DMEM: ProCHO5 media provided the highest VCC in perfusion mode. While the cell growth rate was almost the same in all three media until about the 10th day of culture, temperature shock was performed on the 8th day of culture to increase the yield of production. By applying temperature shift, growth curve for these three media took a different behavior and cell growth continued in a stable trend only in CHO-S-SFM II medium and the quantity of target protein (specific expression) and quality (specific activity) in this medium was higher compared to other media. Taking into account all results, the intensified process in perfusion bioreactor could be a promising approach for the industrial-scale and GMP manufacturing of rh-IFN β-1a as a medicine administered in treatment of multiple sclerosis patients.

Funding

This work was part of IFN301 project which was fully funded and supported by ACTOVERCO Biotech Company.

Data availability

All relevant data can be provided upon request.

Declarations

Conflict of interest

The authors declare no commercial or financial conflict of interest.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. Altamirano C, Paredes C, Illanes A, et al. Strategies for fed-batch cultivation of t-PA producing CHO cells: substitution of glucose and glutamine and rational design of culture medium. J Biotechnol. 2004;110:171–179. doi: 10.1016/j.jbiotec.2004.02.004. [DOI] [PubMed] [Google Scholar]
  2. Arora M. cell culture media: a review. Mater Methods. 2013;3:175. doi: 10.13070/mm.en.3.175. [DOI] [Google Scholar]
  3. Bandaranayake AD, Almo SC. Recent advances in mammalian protein production. FEBS Lett. 2014;588:253–260. doi: 10.1016/j.febslet.2013.11.035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bedoya-López A, Estrada K, Sanchez-Flores A, et al. Effect of temperature downshift on the transcriptomic responses of Chinese hamster ovary cells using recombinant human tissue plasminogen activator production culture. PLoS ONE. 2016;11:0151529. doi: 10.1371/journal.pone.0151529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bustin SA. Absolute quantification of mRNA using real-time reverse transcription polymerase chain reaction assays. J Mol Endocrinol. 2000;25:169–193. doi: 10.1677/jme.0.0250169. [DOI] [PubMed] [Google Scholar]
  6. Chotteau V. Fed-batch or perfusion for the production of biopharmaceuticals by animal cell cultivation. Stockholm: Workshop Securing Quality in Bioanalysis and Production; 2009. [Google Scholar]
  7. Chuppa S, Tsai YS, Yoon S, et al. Fermentor temperature as a tool for control of high-density perfusion cultures of mammalian cells. Biotechnol Bioeng. 1997;55:328–338. doi: 10.1002/(SICI)1097-0290(19970720)55:2<328::AID-BIT10>3.0.CO;2-D. [DOI] [PubMed] [Google Scholar]
  8. Fox SF, et al. Maximizing interferon-γ production by chinese hamster ovary cells through temperature shift optimization: experimental and modeling. Biotechnol Bioeng. 2003;85:177–184. doi: 10.1002/bit.10861. [DOI] [PubMed] [Google Scholar]
  9. Fox SR, et al. A detailed understanding of the enhanced hypothermic productivity of interferon-gamma by Chinese-hamster ovary cells. Biotechnol Appl Biochem. 2005;41:255–264. doi: 10.1042/BA20040066. [DOI] [PubMed] [Google Scholar]
  10. Frerichs KU. Suppression of protein synthesis in brain during hibernation involves inhibition of protein initiation and elongation. Proc Natl Acad Sci USA. 1998;95:14511–14516. doi: 10.1073/pnas.95.24.14511. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Furukawa K, Ohsuye K. Effect of culture temperature on a recombinant CHO cell line producing a C-terminal α-amidating enzyme. Cytotechnology. 1998;26:153–164. doi: 10.1023/A:1007934216507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Gessani S, Conti L, Del Cornò M, et al. Type I interferons as regulators of human antigen presenting cell functions. Toxins. 2014;6:1696–1723. doi: 10.3390/toxins6061696. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Karst DJ, Steinebach F, Morbidelli M. Continuous integrated manufacturing of therapeutic proteins. Curr Opin Biotechnol. 2018;53:76–84. doi: 10.1016/j.copbio.2017.12.015. [DOI] [PubMed] [Google Scholar]
  14. Kumar N, Gammell P, Clynes M. Proliferation control strategies to improve productivity and survival during CHO based production culture. Cytotechnology. 2007;53:33–46. doi: 10.1007/s10616-007-9047-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Kumar N, Gammell P, Meleady P. Differential protein expression following low temperature culture of suspension CHO-K1 cells. BMC Biotechnol. 2008;8:42. doi: 10.1186/1472-6750-8-42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Leclerc GJ, Leclerc GM, Barredo JC (2002) Real-time RT-PCR analysis of mRNA decay: half-life of Beta-actin mRNA in human leukemia CCRF-CEM and Nalm-6 cell lines. Cancer Cell Int 2:1. 10.1186/1475-2867-2-1 [DOI] [PMC free article] [PubMed]
  17. Li F, Vijayasankaran N, Shen A, et al. Cell culture processes for monoclonal antibody production. Mabs. 2010;2:466–477. doi: 10.4161/mabs.2.5.12720. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Lin H, Leighty RW, Godfrey S, et al. Principles and approach to developing mammalian cell culture media for high cell density perfusion process leveraging established fed-batch media. Biotechnol Prog. 2017;33:891–901. doi: 10.1002/btpr.2472. [DOI] [PubMed] [Google Scholar]
  19. Masterton RJ, Smales CM. The impact of process temperature on mammalian cell lines and the implications for the production of recombinant proteins in CHO cells. Pharm Bioprocess. 2014;2:49–61. doi: 10.4155/pbp.14.3. [DOI] [Google Scholar]
  20. Mohsenzadegan M, Fayazi MR, Abdolmaleki M, et al. Direct immunomodulatory influence of IFN-β on human astrocytoma cells. Immunopharmacol Immunotoxicol. 2015 doi: 10.3109/08923973.2015.1014559. [DOI] [PubMed] [Google Scholar]
  21. Ohnishi T, Wang X, Ohnishi K, et al. p53dependent induction of WAF1 by cold shock in human glioblastoma cells. Oncogene. 1998;16:1507–1511. doi: 10.1038/sj.onc.1201663. [DOI] [PubMed] [Google Scholar]
  22. Ozturk S, Hu WS. Cell culture technology for pharmaceutical and cell-based therapies. Taylor Francis. 2005 doi: 10.1201/9780849351068. [DOI] [Google Scholar]
  23. Palomares LA, Estrada-Mondaca S, Ramírez OT. Production of recombinant proteins, challenges and solutions. Methods Mol Biol. 2004;267:15–52. doi: 10.1385/1-59259-774-2:015. [DOI] [PubMed] [Google Scholar]
  24. Pereira S, Kildegaard HF, Andersen MR. Impact of CHO metabolism on cell growth and protein production: an overview of toxic and inhibiting metabolites and nutrients. Biotechnol J. 2018;13:1700499. doi: 10.1002/biot.201700499. [DOI] [PubMed] [Google Scholar]
  25. Pollock J, Ho SV, Farid SS. Fed-batch and perfusion culture processes: economic, environmental, and operational feasibility under uncertainty. Biotechnol Bioeng. 2010;110:206–219. doi: 10.1002/bit.24608. [DOI] [PubMed] [Google Scholar]
  26. Puente-Massaguer E, Badiella L, Gutiérrez-Granados S, et al. A statistical approach to improve compound screening in cell culture media. Eng Life Sci. 2019;19:315–327. doi: 10.1002/elsc.201800168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Reuveny S, Velez D, Miller L, et al. Factors affecting cell growth and monoclonal antibody production in stirred reactors. J Immunol Methods. 1986;86:53–59. doi: 10.1016/0022-1759(86)90264-4. [DOI] [PubMed] [Google Scholar]
  28. Rodriguez J, Spearman M, Huzel N, et al. Enhanced production of monomeric interferon-β by CHO cells through the control of culture conditions. Biotechnol Prog. 2008;21:22–30. doi: 10.1021/bp049807b. [DOI] [PubMed] [Google Scholar]
  29. Rodriguez J, Spearman M, Tharmalingam T, et al. High productivity of human recombinant beta-interferon from a low-temperature perfusion culture. J Biotechnol. 2010;150:509–518. doi: 10.1016/j.jbiotec.2010.09.959. [DOI] [PubMed] [Google Scholar]
  30. Roobol A, Roobol J, Carden MJ. ATR (ataxia telangiectasia mutated- and Rad3-related kinase) is activated by mild hypothermia in mammalian cells and subsequently activates p53. Biochem J. 2011;435:499–508. doi: 10.1042/BJ20101303. [DOI] [PubMed] [Google Scholar]
  31. Ross J. mRNA stability in mammalian cells. Microbiol Rev. 1995;59:423–450. doi: 10.1128/mr.59.3.423-450.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Shani N, Isolini D, Marzohl D, et al. Evaluation of a new culture medium for the enumeration and isolation of Streptococcus salivarius subsp. thermophilus from cheese. Food Microbiol. 2021;95:103672. doi: 10.1016/j.fm.2020.103672. [DOI] [PubMed] [Google Scholar]
  33. Spearman M, Rodriguez J, Huzel N, et al. Effect of culture conditions on glycosylation of recombinant beta-interferon in CHO cells. Cell Technol Cell Prod. 2007 doi: 10.1007/978-1-4020-5476-1_11. [DOI] [Google Scholar]
  34. Tharmalingam T, Sunley K, Butler M. High yields of monomeric recombinant β interferon from macroporous microcarrier cultures under hypothermic conditions. Biotechnol Prog. 2008;24:832–838. doi: 10.1002/btpr.8. [DOI] [PubMed] [Google Scholar]
  35. van Mastrigt O, Abee T, Lillevang SK, et al. Quantitative physiology and aroma formation of a dairy Lactococcus lactis at near-zero growth rates. Food Microbial. 2018;73:216–226. doi: 10.1016/j.fm.2018.01.027. [DOI] [PubMed] [Google Scholar]
  36. Whitford WG. Fed-batch mammalian cell culture in bioproduction. BioProcess Int. 2006;4:30–40. [Google Scholar]
  37. Xu J, Tang P, Yongky A, et al. Systematic development of temperature shift strategies for Chinese hamster ovary cells based on short duration cultures and kinetic modeling. Mabs. 2019;11:191–204. doi: 10.1080/19420862.2018.1525262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Yang H. Selection of culture media for human and rabbit corneal epithelia. Chin J Ophthalmol. 1991;27:351–353. [PubMed] [Google Scholar]
  39. Yang Z, Xiong HR. Culture conditions and types of growth media for mammalian cells. Biomed Tissue Cult. 2012 doi: 10.5772/52301. [DOI] [Google Scholar]
  40. Yoon SK, Song JY, Lee GM. Effect of low culture temperature on specific productivity, transcription level, and heterogeneity of erythropoietin in Chinese hamster ovary cells. Biotechnol Bioeng. 2003;82:289–298. doi: 10.1002/bit.10566. [DOI] [PubMed] [Google Scholar]
  41. Yusufi F, Lakshmanan M, Ho YS, et al. Mammalian systems biotechnology reveals global cellular adaptations in a recombinant CHO cell line. Cell Syst. 2017;4:530–542. doi: 10.1016/j.cels.2017.04.009. [DOI] [PubMed] [Google Scholar]
  42. Zoro B, Tait T (2017) Development of a novel automated perfusion mini bioreactor ‘ambr® 250 perfusion in Integrated Continuous Biomanufacturing III, Suzanne Farid, University College London, United Kingdom Chetan Goudar, Amgen, USA Paula Alves, IBET, Portugal Veena Warikoo, Axcella Health, Inc., USA Eds, ECI Symposium Series. https://dc.engconfintl.org/biomanufact_iii/43

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

All relevant data can be provided upon request.

Articles from Cytotechnology are provided here courtesy of Springer Science+Business Media B.V.

RESOURCES