Scholarly article on topic 'Metabolic engineering of CHO cells to alter lactate metabolism during fed-batch cultures'

Metabolic engineering of CHO cells to alter lactate metabolism during fed-batch cultures Academic research paper on "Biological sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Journal of Biotechnology
OECD Field of science
Keywords
{"CHO cells" / "Pyruvate carboxylase" / "Metabolic engineering" / Fed-batch / "Monoclonal antibody"}

Abstract of research paper on Biological sciences, author of scientific article — Cécile Toussaint, Olivier Henry, Yves Durocher

Abstract Recombinant yeast pyruvate carboxylase (PYC2) expression was previously shown to be an effective metabolic engineering strategy for reducing lactate formation in a number of relevant mammalian cell lines, but, in the case of CHO cells, did not consistently lead to significant improvement in terms of cell growth, product titer and energy metabolism efficiency. In the present study, we report on the establishment of a PYC2-expressing CHO cell line producing a monoclonal antibody and displaying a significantly altered lactate metabolism compared to its parental line. All clones exhibiting strong PYC2 expression were shown to experience a significant and systematic metabolic shift toward lactate consumption, as well as a prolonged exponential growth phase leading to an increased maximum cell concentration and volumetric product titer. Of salient interest, PYC2-expressing CHO cells were shown to maintain a highly efficient metabolism in fed-batch cultures, even when exposed to high glucose levels, thereby alleviating the need of controlling nutrient at low levels and the potential negative impact of such strategy on product glycosylation. In bioreactor operated in fed-batch mode, the higher maximum cell density achieved with the PYC2 clone led to a net gain (20%) in final volumetric productivity.

Academic research paper on topic "Metabolic engineering of CHO cells to alter lactate metabolism during fed-batch cultures"

Accepted Manuscript

Journal of

BIOTECHNOLOGY

Title: Metabolic engineering of CHO cells to alter lactate metabolism during fed-batch cultures

Author: Cécile Toussaint Olivier Henry Yves Durocher

PII: DOI:

Reference:

S0168-1656(15)30182-6

http://dx.doi.Org/doi:10.1016/j.jbiotec.2015.11.010 BIOTEC 7303

To appear in:

Journal of Biotechnology

Received date: Revised date: Accepted date:

13-8-2015 3-11-2015 13-11-2015

Please cite this article as: Toussaint, Cecile, Henry, Olivier, Durocher, Yves, Metabolic engineering of CHO cells to alter lactate metabolism during fed-batch cultures.Journal of Biotechnology http://dx.doi.org/10.1016/j.jbiotec.2015.11.010

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Metabolic engineering of CHO cells to alter lactate metabolism during fed-batch cultures

Cécile Toussaint a,b , Olivier Henry c , Yves Durocher a'b'*

a Département de Biochimie et médecine moléculaire, Université de Montréal, Montréal,

Québec H3C 3J7, Canada

cécile.toussaint@umontreal.ca

b Life Sciences, NRC Human Health Therapeutics Portfolio, Building Montreal-Royalmount, National Research Council Canada, Montreal, QC H4P 2R2, Canada yves.durocher@nrc-cnrc.gc.ca

c Département de Génie Chimique, École Polytechnique de Montréal, C.P. 6079, Succ.

Centre-ville, Montréal, Québec H3C 3A7, Canada.

o.henry@polymtl.ca

Highlights

• PYC2 expression in CHO cells induced a systematic metabolic shift toward lactate consumption in both batch and fed-batch modes, even under elevated nutrient concentrations.

• A feeding strategy employing a concentrated nutrient solution was implemented to take advantage of the improved phenotype of PYC2-expressing CHO cells, leading to higher volumetric titer and without significant waste metabolite accumulation.

• PYC2 clones exhibited a much more efficient metabolism characterized by lower lactate/glucose and ammonia/glutamine ratios.

Abstract:

Recombinant yeast pyruvate carboxylase (PYC2) expression was previously shown to be an effective metabolic engineering strategy for reducing lactate formation in a number of relevant mammalian cell lines, but, in the case of CHO cells, did not consistently lead to significant improvement in terms of cell growth, product titer and energy metabolism efficiency. In the present study, we report on the establishment of a PYC2-expressing CHO cell line producing a monoclonal antibody and displaying a significantly altered lactate metabolism compared to its parental line. All clones exhibiting strong PYC2 expression were shown to experience a significant and systematic metabolic shift toward lactate consumption, as well as a prolonged exponential growth phase leading to an increased maximum cell concentration and volumetric product titer. Of salient interest, PYC2-expressing CHO cells were shown to maintain a highly efficient metabolism in fed-batch cultures, even when exposed to high glucose levels, thereby alleviating the need of controlling nutrient at low levels and the potential negative impact of such strategy on product glycosylation. In bioreactor operated in fed-batch mode, the higher maximum cell density achieved with the PYC2 clone led to a net gain (20 %) in final volumetric productivity.

Keywords: CHO cells, Pyruvate carboxylase, metabolic engineering, fed-batch, monoclonal antibody

1. Introduction

Whether to support new therapeutic antibody development or to respond to market needs, the ability to cost-efficiently generate high yield and high quality of biotherapeutics is critical for the biopharmaceutical industry, owing to the great pressure to lower the cost and shorten the period between drug discovery and product manufacturing. The development and implementation of high-yield fed-batch strategies has contributed greatly at increasing the productivity of mammalian cell cultures (Bibila and Robinson, 1995; Wlaschin and Hu, 2006; Wurm, 2004). In fed-batch processes, the prolonged culture time is often associated with significant lactate and ammonia accumulation in the culture medium over time, which can have detrimental impacts on cell growth and product quality. Continuous cell lines, like CHO cells, show a deregulated glucose metabolism associated with high lactate production that can cause medium acidification or undesired osmolality changes due to alkali addition for culture pH control (Cruz et al., 2000; Lao and Toth, 1997; Omasa et al., 1991; Ozturk et al., 1992). Ammonia is well-known to alter the glycosylation patterns of monoclonal antibodies and other glycoproteins (Andersen and Goochee, 1995; Borys et al., 1994; Chen and Harcum, 2006; Gawlitzek et al., 2000; Yang and Butler, 2002). Appropriate glycosylation of monoclonal antibodies and recombinant proteins is necessary to ensure their biological function and their use as therapeutics. Thus, maintaining consistent and comparable product quality in high cell density cultures remains a critical challenge.

Several waste reduction strategies have been proposed using either metabolic or process

engineering approaches. This includes the use of nutrients replacement strategies, for example

the substitution of glucose with galactose (Altamirano et al., 2006; Altamirano et al., 2000;

Altamirano et al., 2004) or pyruvate (Genzel et al., 2005) or substituting glutamine with

asparagine (Kurano et al., 1990) or glutamate (Altamirano et al., 2000; Hassel and Butler,

1990). While effective at reducing waste metabolite formation, these substitutions are often

associated with a concomitant reduction in cellular growth. Reduced lactate and ammonia production during fed-batch can be achieved by maintaining glucose and glutamine at very low levels (Omasa et al., 1992; Zhou et al., 1995). Under low nutrient availability, the cell metabolism is shifted toward a more efficient metabolic state characterized by a lower molar ratio of lactate/glucose (Chee Furng Wong et al., 2005; Maranga and Goochee, 2006). However, this requires the constant monitoring of nutrient levels and feed rate adjustments since non-limiting concentrations must be maintained at all time during the culture. Even short period of glucose starvation were shown to alter the glycosylation of the product (Xie et al., 1997). Moreover, it has been demonstrated that low glucose concentrations (<15mM) lead to reduced site occupancy and decreased galactosylation and sialylation of antibodies (Liu et al., 2014). A dynamic fed-batch strategy to maintain low glutamine (0.1 - 0.3 mM) and glucose (0.35-0.70 mM) concentrations was also shown to decrease complex-type glycan formation and sialic acid content (Chee Furng Wong et al., 2005). These results emphasize the need to optimize both product titer and quality concurrently.

Different metabolic engineering approaches have also been investigated to generate cell lines

with improved metabolic characteristics. The amplification of the glutamine synthetase gene

in CHO and myeloma cells allowed growth in glutamine-free medium, thereby significantly

reducing ammonia formation (Paredes et al., 1999; Zhang et al., 2006). Partial disruption of

the gene encoding lactate dehydrogenase A by the means of homologous recombination in

hybridoma cells (Chen et al., 2001) or siRNA in CHO cells (Jeong et al., 2006; Kim and Lee,

2007a) was shown to greatly lower the production of lactate, as well as glucose consumption.

Studies by Chen et al. (2001) and Kim and Lee (2007a) on the down regulation of LDHA led

to increased final product concentrations, although cell growth remained unaffected. In

contrast, Jeong et al. (2006) observed an increased cell growth rate, but no significant impact

on tPA production unless downregulation of LDH was combined with the overexpression of

glycerol-3-phosphate dehydrogenase. A recent study demonstrated that the simultaneous downregulation of LDH and pyruvate dehydrogenase kinase can significantly reduce lactate accumulation and increase the volumetric antibody production, without impairing cell growth (Zhou et al., 2011). The expression of anti-apoptotic genes was also found to significantly alter the lactate metabolism of CHO cells, inducing a metabolic shift toward lactate consumption and increasing the final antibody titer by 40% (Dorai et al., 2009). Another approach consisted in the expression of the cytosolic yeast pyruvate carboxylase 2 (PYC2) gene in BHK (Irani et al., 1999) and HEK293 cells (Elias et al., 2003; Henry and Durocher, 2011; Vallée et al., 2013). In all cases, lactate formation was significantly decreased and, depending on the cell type and prevailing culture conditions, translated into improvement of either productivity or cell growth, but not both. In the case of CHO cells, two strategies were explored. First, the insertion of the human pyruvate carboxylase (hPC) gene in DG44 cells, which led to a slight decrease in lactate production (Kim and Lee, 2007b), as this enzyme is expressed in the mitochondria and is therefore not competing directly with LDH for pyruvate conversion. Second, one study reported on the expression of PYC2 in CHO cells, which allowed to decrease the specific lactate production (Fogolrn et al., 2004). However, the molar ratio of lactate/glucose remained high (~ 1.5) and relatively unchanged compared to the untransformed cells (~ 1.7), indicating that the reduction in lactate accumulation was mostly due to a decrease in glucose uptake rate. Moreover, while prolonged cell viability was observed, this was at the expense of the maximum cell density, which exhibited a 2-fold reduction in batch culture. Recently, PYC2 overexpression in CHO cells was shown to prolong culture longevity and reduce the lactate/glucose ratio by 25%, but at the expense of a 50 % reduction in the cell specific antibody production rate and an overall decrease in antibody titer (Wilkens and Gerdtzen, 2015).

In the present work, we show the establishment of a PYC2-expressing CHO cell line with a greatly improved metabolic efficiency that translated into reduced lactate accumulation without impairing the cells' specific productivity. All the PYC2- expressing clones that were generated exhibited a prolonged exponential growth phase leading to increases in maximum cell concentration and volumetric productivity. We also demonstrate that the PYC2-expressing cells can maintain their improved phenotype during fed-batch cultures employing concentrated nutrient solutions. In fed-batch bioreactor cultures, this resulted in further significant increases in terms of maximum cell density and final product titer.

2. Materials and Methods

2.1. Stable cell line development

The CHO-EG2 clone 1A7 stably expressing a chimeric human-llama heavy chain monoclonal

antibody (EG2) against epidermal growth factor receptor (EGFR) (Bell et al., 2010) was

established from the CHO-DUXB11 cell line (Agrawal et al., 2012). The CHO-EG2-PYC2

clone was obtained by transfection of the CHO-EG2 cell line with the pTT18 vector encoding

the yeast pyruvate carboxylase 2 gene. After transfection, hygromicin B was added to the cell

culture at a final concentration of 900^g/mL. After two weeks of selection, the pool of

hygromycin-resistant cells was cloned by limiting dilution in 96-well plate. After 2 weeks,

colonies were expanded and screened for their PYC2 expression level by western blot. Since

the PYC2 enzyme is naturally biotinylated, it allows its detection using HRP-conjugated

streptavidin (Sigma). Briefly, cells pellets were incubated on ice 20 minutes in a lysis buffer

containing 50 mM HEPES pH 7.4, 150 mM NaCl, 1% Thesit® (Roche), 0.5% Na

Deoxycholate, 1x complete-protease inhibitor cocktail (Roche). Cell lysates were centrifuged

for 20 min at 12,000 g at 4°C. Supernatants were incubated 5 min at 70°C in SDS sample buffer containing 5 mM dithiothreitol. Samples were electrophoresed on an SDS polyacrylamide gel electrophoresis (PAGE) and transferred onto a nitrocellulose membrane (Schleicher & Schuell). Membranes were probed with HRP-conjugated streptavidin (Sigma). An anti-GAPDH rabbit polyclonal antibody (Sigma) was utilized for detection of the GAPDH to control total protein load. As a secondary antibody, an anti-rabbit polyclonal antibody HRP-conjugated (sigma) was employed. The blots were developed using a chemiluminescence kit (Boehringer Mannheim, Germany) and visualized with the Kodak imager system 440CF.

2.2. Fed-batch culture in shake flask for clone evaluation

Eleven PYC2-positive clones and 4 PYC2-negatives clones were cultivated in 250 mL shake-flasks. After 4 days of cultivation, glucose was added daily in order to maintain a concentration of 20 mM. Lactate formation was measured daily for phenotypic screening to confirm that the change in lactate metabolism was the results of PYC2 over-expression rather than clonal effect.

2.3. PYC2 Cytosolic expression

Three clones were selected according to their stable expression of PYC2. In order to confirm cytosolic PYC2 expression, the cytosols of these clones were isolated using mitochondria isolation kit for cultured cells (Thermo Scientific- cat#89874). Their PYC2 expression level was assessed by western blot using HRP-conjugated streptavidin (Sigma) as described before.

2.4. Medium and culture conditions

PowerCHO2 chemically defined medium (Lonza) supplemented with 8 mM glutamine, 0.1%

Kolliphor® (Sigma) and 2% (v/v) anti clumping agent (Gibco) was used for stable cell line

development and batch shake-flasks experiments. Cultures were inoculated at a cell concentration of 0.2x106 cells/mL and were grown at 37°C and 5% CO2 under constant agitation (120 rpm).

2.5. pH monitoring in 6 well plates

Cell cultures were inoculated at a cell concentration of 0.2x106 cells/mL and grown at 37°C and 5% CO2 under constant agitation (120 rpm) in 6-well plates with integrated pH sensors (HydroDish®, PreSens-Precision Sensing GmbH, Germany).

2.6. Fed-batch cultivation

For fed-batch cultivation in shake flask or bioreactor, cells were seeded at a target cell concentration of 0.2x106 cells/mL in BalanCD (Irvine) chemically defined medium supplemented with 8 mM glutamine, 0.1% Kolliphor® and 2% (v/v) anti clumping agent (Gibco). Samples were taken daily to measure the cell density and nutrient/metabolite concentrations. Once the analyses completed, addition of CHO CD EfficientFeed™ B (Gibco) supplemented with 40 mM of glutamine was performed on a daily basis. The daily feeding volume is calculated according to the cell specific consumption of glucose:

(R^IVCDpredicted )"(Ct XV - Ctarget XV) Feed Volume=-

Where R denotes the specific nutrient consumption rate. Ct and Ctarget are the nutrient concentration and the desired nutrient concentration, respectively. V denotes the volume of the vessel and Cfeed represents the nutrient concentration in the feed.

2.7. Bioreactor

Bioreactor experiments were performed in 2 L stirred bioreactor (Labfors 4 cell, Infors HT, Switzerland) operated at a 0.6 L starting volume. Dissolved oxygen was maintained at 50% of air saturation by surface aeration using a gas mixture of oxygen/nitrogen/air with a gas flow set at 100 mL/min. When necessary, pure oxygen was sparged at 2 mL/min with an on/off controller. The temperature was maintained at 37°C with a double-wall water-jacketed. The stirring speed started at 80 rpm and was gradually increased during culture to reduce cell clumping and optimize gas transfer, until a maximum of 120 rpm. Culture pH was controlled at 7.0 by a surface gassing of CO2 or 1 M sodium hydroxide addition. Addition of feeding medium was operated with a digital tubing pump delivering the desired feed volume at a flow rate of 280 |iL/min.

2.8. Analytical methods

2.8.1. Cell count and viability, glucose, lactate, ammonium, glutamine quantification

Cell number and viability were assessed by trypan blue exclusion and counting on a hemocytometer. Samples were centrifuged and the supernatants were frozen at -80°C for subsequent analysis. The glucose, lactate and glutamine concentrations in supernatants were measured with YSI 2700 Select biochemistry analyzer (Yellow Spring, USA). Ammonia concentration was measured by the VITROS 350 (Orthoclinical Diagnostics, USA).

2.8.2. Antibody quantification

Antibody quantification in supernatants was determined by high performance liquid

chromatography (WATERS Corporation, Milford, MA) using a protein A cartridge

(POROS® A20 column, 2.1 mmD x 30 mmH, 104 pL, Invitrogen, Grand Island, NY). Samples were filtered by centrifugation at 8000g-11000g for 3 minutes using NANOSEP MF GHP 0.45^m centrifugal devices (PALL Life Sciences) prior to being injected on the column at a flow rate of 2 mL/min (PBS) and elution was performed using 0.15M NaCl, pH 2.0. UV detection was done at 280 nm.

3. Results

3.1. Generation of CHO clones overexpressing PYC2

The pTT18 vector encoding the yeast pyruvate carboxylase 2 gene was transfected in CHO

cells stably expressing a chimeric human-llama heavy chain monoclonal antibody (EG2). 129

individual clones were assessed for their level of PYC2 expression by western blot (Figure

1A). For the parental cell line (Lane 1), the observed signal corresponds to the presence of the

endogenous pyruvate carboxylase (PC) naturally expressed in the mitochondria of cells and

which bears the same apparent molecular weight. PYC2-expressing clones (e.g., lanes 2, 4, 7

and 24 in Figure 1A) all displayed a stronger signal reflecting the combined expression of PC

and PYC2. Clones transfected with the pTT18 vector but showing a signal similar to the

parental cells (e.g., lanes 3, 5, 6, 8-23 and 25 in Figure 1A) were considered as negative.

Finally, 11 clones that exhibited significant PYC2 expression were selected and used for

further analysis (Figure 1B). Given that PYC2 is expected to be expressed in the cytosol (in

contrast to endogenous PC that is localized in the mitochondria), a subcellular fractionation

was performed on three PYC2-positive clones (clones 1, 7 and 11 from Figure 1C) that were

selected on the basis of their stable PYC2 expression for over 2 months in culture. As shown

in Figure 1D, western blot analysis revealed a signal corresponding to PYC2 that was indeed

specific to the transfected clones and confirmed its proper expression in the cytosol. As expected, GAPDH expression was observed in all the cytosolic fractions.

3.2. Phenotypic evaluation of PYC2 clones

To rapidly assess the impact of PYC2 expression on cell growth and metabolism, 11 PYC2-positive clones were cultivated in shake flasks operated in fed-batch mode, whereby glucose was added on a daily basis to maintain a target concentration of 20 mM. Maintaining glucose in excess was used to create conditions favoring an overflow metabolism (i.e. lactate production) and to avoid limitations that could alter lactate metabolism as reported in previous studies (Altamirano et al., 2006; Martinez et al., 2013). To rule out potential effects attributable to clonal variation, the kinetics of cell growth and lactate production of the 11 PYC2-positive clones were evaluated and compared to that of the parental cell line and to 4 PYC2-negative clones that were randomly chosen to serve as negative controls. As shown in Figure 2A and in striking contrast with the negative clones and the parental cells, all the PYC2-expressing clones exhibited the same altered metabolism characterized by a significant shift towards lactate consumption around day 6. Noteworthy, cell growth was not negatively impacted and most of the positive clones even reached slightly higher maximum viable cell concentrations (Figure 2B).

Several studies have previously shown that the culture medium composition can strongly

influence the metabolism of a given cell line, including its pattern of lactate

production/consumption (Altamirano et al., 2006; Luo et al., 2012; Ma et al., 2009; Yuk et al.,

2014; Zagari et al., 2013a). In order to confirm the consistency of the metabolic shift

displayed by the PYC2 clones, cells were cultivated in 5 commonly used commercial media.

As shown in Figure 3A, the final lactate concentrations varied depending on the culture

medium, but were always consistently lower in the case of the PYC2-expressing clones

compared to the parental cells. It should be emphasized that the differences observed were not the result of significant changes in cell densities, as the calculated cell specific lactate production rates were also consistently lower for PYC2 clones compared to the parental cell line (data not shown). Of salient interest, significant reductions in final ammonia concentrations (up to 50 %) were also observed for PYC2 clones in three of the five commercial media formulation tested (Figure 3B).

Lactate accumulation and the consequent medium acidification in uncontrolled flask cultures can be detrimental to the cells and/or the product quality. We therefore investigated the impact of PYC2 expression on the evolution of pH. To that end, cells were cultivated in batch mode in a 6-well plate and pH was monitored using a sensor dish reader (SensorDish®). The recorded pH signals are presented in Figure 4 and clearly demonstrate that the acidification of the medium is significantly reduced in cultures performed with PYC2 clones.

3.3. Batch and fed batch cultivations in shake flasks

In order to assess the potential of PYC2 clones for fed-batch applications, cells were

cultivated in shake flasks under batch and fed-batch modes using the BalanCD medium,

which yielded the highest cell density in shake flask batch cultures. Clone #11 (Figure 1) was

used in all the subsequent experiments in shake flask and bioreactor cultures. The resulting

cell density profiles are shown in Figure 5A for both the parental and PYC2-expressing cells.

When cultivated in batch mode, the growth curves differed markedly between both cell lines.

The PYC2 cell line reached a maximum cell concentration of about 1x107 cells/mL, a nearly

two-fold increase compared with the parental cells. The maximum cell specific growth rates

were also fairly different for the two cell lines (0.033 and 0.023 h-1 for the PYC2 and the

parental clones, respectively) and the exponential phase of the PYC2 clone was extended by

more than 24 hours. Upon reaching their maximum cell density on day 6, the PYC2 cells

exhibited a sharp decrease in viable cell concentration thereafter (Figure 5B). This sudden drop in cell viability was most likely due to a key nutrient depletion given the high cell density reached in these batch cultures. When grown in fed-batch, the PYC2 cells had a similar cell specific growth rate (0.032h-1), but the feeding strategy allowed to maintain high cell densities over a prolonged period of time, as the cell viability decreased below 85 % only after day 7. In contrast, the same feeding regime had only marginal impact on the growth and viability of the parental cells (Figures 5A & 5B) compared with the batch mode. Batch cultures were characterized by a nearly total depletion of glucose by day 8 for both PYC2 and parental cells (Figure 5C), while the fed-batch strategy allowed to maintain glucose around the target concentration of35 mM. Noteworthy, despite those elevated glucose concentrations, the PYC2 clone exhibited the same altered lactate metabolism witnessed in batch (Figure 5D). A comparison of the main cell specific consumption and production rates evaluated during the exponential and the stationary phases is presented in Table 1. In batch and fed-batch cultures, only slight reduction in the glucose specific consumption rates were observed for PYC2 cells, but marked differences were noted for lactate as the specific production was reduced by 50% during the exponential phase. The differences were even greater during the stationary phase, as lactate was significantly consumed by PYC2 cells, unlike in the parental cell line (Table 1). As evident from the calculated lactate/glucose ratios, the PYC2 clone exhibited a much more efficient glucose utilization.

In batch cultures, glutamine depletion occurred rapidly (by day 4), but the average cell

specific glutamine consumption rates were not significantly different for the two cell lines

(Figure 5E & Table 1). The supplementation of glutamine in fed-batch cultures led to an

increased consumption rate and greater ammonia accumulation, particularly for the PYC2

clone due to the greater cell density and the prolonged culture. While the specific productions

were quite similar during the exponential phase, both cell lines showed a decreased

production of ammonia per cell during the stationary phase (Table 1). Whether the cultures were performed in batch or fed-batch modes, the final product titer in cultures with PYC2-expressing cells was 35% greater compared to the parental cell line (Figure 7). This augmentation can be attributed to both the increased cell density and the prolonged culture longevity associated with PYC2 expression. The cell specific productivity of PYC2-expressing cells was comparable in batch and fed-batch modes (1.91 ± 0.40 and 2.02 ± 0.44 pg/cell d, respectively), while that of the parental clone increased from 1.69 ± 0.35 pg/cell d in batch to 2.69 ± 0.54 pg/cell d in fed-batch. Compared to the parental cell line, the slightly lower cell specific productivity of the PYC2 clone is largely compensated by the increased cell growth.

3.4. Fed-batch cultivation in bioreactor

The sharp and sudden viability decrease witnessed in shake flasks suggested that these

uninstrumented culture conditions could not support the high cell densities reached by PYC2

cells (around 1x107 cells/mL). Cells were thus cultivated under fed-batch mode using fully

controlled bioreactors so as to provide better regulated conditions in terms of pH and oxygen.

The feeding strategy consisted in the daily addition of a concentrated solution (CHO CD

EfficientFeed B) supplemented with 40 mM of glutamine. The resulting cell density profiles

are shown in Figure 6A. The PYC2 cell line reached a maximum cell concentration of about

1.3x107 cells/mL, a more than 2-fold increase compared with the parental cells. The

maximum cell specific growth rates were fairly similar for the two cell lines (0.023 and 0.024

h-1 for the PYC2 and the parental clones, respectively), but the exponential growth phase was

extended by 3 days in the case of the PYC2 clone. Upon reaching their maximum in

concentration between days 9 and 12 (13x106 cells/mL), there was a sharp decrease in viable

cell density, similar to the trend observed in small-scale cultures (Figure 6B). Since the

dissolved oxygen set-point was successfully maintained at around 50% throughout the culture, this was likely due to some other nutrient limitation.

As was observed in shake flask cultures, the expression of the PYC2 gene led to a significant alteration of the cell metabolism, with lactate being consumed after 7 days of cultivation despite elevated glucose concentration (Figure 6C & 6D). The greater nutrient utilization efficiency of the PYC2 cell line was even observed during the exponential phase, as the cell specific lactate production rate was reduced by 80 % for a comparable glucose uptake rate (Table 2). PYC2 expression did not significantly impact the glutamine metabolism, and, as a result, the cultures achieved similar final levels of ammonia (Figure 6E & 6F), although the specific ammonia production rate of the PYC2 clone was lower (Table 2). Consistent with small-scale fed-batch results, the average cell specific antibody production rates was slightly lower for the PYC2 clone (2.21± 0.45 and 2.86± 0.58 pg/cell d for PYC2 and parental cells), but the final product titer was augmented by 28% due to the higher cell density achieved (Figure 7). Of interest, the maximum volumetric titer was reached 4 days earlier in cultures with the PYC2 cells. While the fed-batch strategy brought only a slight antibody volumetric productivity improvement in shake flasks, cultures in bioreactor improved by around 30 to 40% the volumetric production of PYC2 and parental cell lines, respectively.

4. Discussion

The aim of this work was to generate a PYC2-expressing CHO cell line with an improved

metabolic efficiency and demonstrate its potential for the development of a fed-batch process

for antibody production. In any metabolic engineering strategy, clonal variations should be

considered when trying to assess the real impact of gene overexpression. Clone-based studies often fail to establish a relationship between transgene expression and the desired trait. In the specific case of lactate reduction strategies, a variety of other factors may cause differences in metabolic phenotypes, such as changes in the medium composition (Kishishita et al., 2015; Luo et al., 2012; Zagari et al., 2013a). In our study, we have initially compared the kinetics of growth and metabolite formation of 11 PYC2-positive, 4 PYC2-negative and the parental clones. The fact that all 11 PYC2-positive clones exhibited comparable phenotypes when tested in different media formulation provided strong evidence that the differences observed were due to PYC2 expression and not to clonal variations.

The PYC2 expressing CHO cells that were generated displayed a much more efficient

metabolism characterized by a lower lactate/glucose ratio, especially in fed-batch cultures

using an enriched nutrient solution. More importantly, the slightly lower cell specific antibody

production of the PYC2 clone is largely compensated by the increase in maximum cell

density, translating into a net gain in volumetric productivity. PYC2 expressing cells

maintained their distinctive phenotype, even when exposed to high glucose levels during fed-

batch cultures. In addition, the cell specific productivity is not significantly altered by the

metabolic modification, as was previously observed when the same gene was overexpressed

in HEK293 cells producing interferon in batch (Henry and Durocher, 2011) or fed-batch

(Vallée et al., 2013) cultures. In contrast with our results, an apparent two fold increase in

specific productivity was observed in PYC2 transfected BHK-21 cells producing recombinant

EPO, although these cells were cultivated in perfusion mode under limiting glucose

conditions (Irani et al., 2002). However, no positive impacts were observed in terms of

maximum cell density and cell specific growth rate. Similarly, a 2-fold increase of specific

hGM-CSF productivity was noted for CHO cells expressing PYC2 cultivated in batch and

following a temperature switch (Fogolín et al., 2004). The cell specific growth rate and

maximum cell density were reduced, but culture longevity was extended leading to a rise in final product titer. In an earlier study, HEK293 expressing PYC2 showed no significant increase in SEAP titer in spite of increase cell longevity and reduced waste production (Elias et al., 2003). Similarly, a decrease in the specific antibody productivity was found for CHO cells expressing PYC2 , even though the cell growth rate and glucose/lactate ratio were improved (Wilkens and Gerdtzen, 2015). Taken together, these results indicate that the effects of PYC2 expression are dependent on the cell type and the culture conditions. In the present study, the greatest impact of PYC2 expression in CHO cells was on the maximum cell density attained during the cultures. Results from our shake flask and bioreactor cultures in fed-batch mode showed that the increase in cumulative viable cell concentration, combined with the maintenance of antibody specific productivity, resulted into a net increase in the final volumetric titer.

In the transition from small-scale to bioreactor, the metabolism and growth of the PYC2 clone

remained mostly similar, but the product titer increased substantially. This titer improvement

can likely be attributed, at least in part, to tightly regulated conditions in the bioreactor. In the

case of PYC2 cells, regardless of the culture mode (batch or fed-batch) or culture device

(shake flask or bioreactor), the lactate shift seemed to occur systematically 24 hours before

reaching the maximal cell density, as the cells entered the deceleration phase. For the parental

CHO cell line, the net accumulation of lactate ceased when cell growth stopped. In either of

the two cell lines, these results suggest a link between a certain lactate level, growth rate

reduction and lactate production shutdown. In the case of cells expressing PYC2, the enzyme

could have an additive effect leading to a net lactate consumption due to the competition for

pyruvate between this enzyme and lactate dehydrogenase (Irani et al., 1999). Lactate

consumption has been reported to occur under low glucose concentrations (Altamirano et al.,

2004; Cruz et al., 2000; Ozturk et al., 1992) or under non-limiting levels (Luo et al., 2012; Ma

et al., 2009; Mulukutla et al., 2012; Ozturk et al., 1997; Pascoe et al., 2007). The co-consumption of lactate and glucose was identified as a highly desirable trait for industrial cell lines, since it is typically associated with increased culture longevity and greater product yields (Le et al., 2012). This phenomenon and its underlying physiological mechanisms are still not fully understood and, consequently, not yet well-controlled. Recently, some studies have suggested that the copper level in the culture medium could be a trigger of lactate consumption (Luo et al., 2012; Yuk et al., 2014), but other factors have also been associated to this metabolic shift, including an increase in oxygen consumption, a reduction of glycolysis and glutaminolysis, as well as cell growth slowdown (Luo et al., 2012; Ma et al., 2009; Mulukutla et al., 2012; Pascoe et al., 2007).

Studies on BHK-21 expressing PYC2 have reported increases in oxygen uptake and

intracellular ATP pool, pointing to an enhanced TCA activity resulting from this genetic

modification (Irani et al., 1999). More recently, 13C metabolic flux analysis conducted on

HEK293-PYC2 expressing cells confirmed an increase in the pyruvate flux to oxaloacetate

(Henry and Durocher, 2011), although this intermediate metabolite can subsequently have

multiple fates. First, in the cytosol, the latter can be accumulated or could be converted into

malate through cytosolic malate dehydrogenase (MDH) with the concomitant release of

NAD+. Then the malate can be recycled into pyruvate via cytosolic malic enzyme or enter the

mitochondria via the malate/aspartate shuttle to be completely oxidized in the TCA. A

previous study on CHO cells expressing mitochondrial MDH II reported enhanced glycolysis,

reduced lactate secretion and elevated intracellular ATP/NADH levels along with antibody

titer improvement (Chong et al., 2010). These findings suggest a critical role of the redox

balance in the regulation of the glycolytic flux and the proportion of pyruvate entering the

TCA. The capacity of the LDH reaction to regenerate NAD+ allows to maintain high

glycolytic rates, decreasing the availability of cytosolic pyruvate as a substrate for the TCA

cycle. The malate-aspartate shuttle plays a pivotal role in maintaining the redox balance by allowing the regeneration of NAD+ in the cytosol. In CHO cells, overexpression of aralar1, an aspartate/glutamate carrier and critical component of the malate-aspartate shuttle, was shown to promote a metabolic switch to lactate consumption without any reduction of the glycolytic rate (Zagari et al., 2013b) For PYC2 overexpression, it can be speculated that the reduction in the LDH flux is likely compensated by an increase in the malate dehydrogenase flux to maintain the redox balance.

The distinctive phenotype of PYC2-expressing CHO cells can provide additional benefits in the context of the development of a cell culture manufacturing process. First, it is well known that glycosylation pattern of glycoproteins can be negatively altered by elevated waste product accumulation (Andersen and Goochee, 1995; Gawlitzek et al., 2000; Yang and Butler, 2002). Since significant lactate/ammonia accumulation is likely to occur in standard high cell density fed-batch cultures, the lower waste metabolite formation obtained in PYC2 cells could be expected to enhance final product quality and stability in the culture medium, but further work is required to assess the impact of PYC2 overexpression on product quality attributes. Furthermore, this phenotypic change eliminates the need to control glutamine/glucose at low levels, strategies which make process operation more complex and also have adverse effects on the glycosylation profile of the protein of interest (Liu et al., 2014). Second, it is generally recognized that small-scale studies performed in shake flasks during process development may not translate well to bioreactor during scale-up, particularly due to lactate accumulation. Given that the pH changes are damped with PYC2-expressing cells, this represents another significant advantage to streamline the optimization of culture conditions.

5. Conclusion

We have demonstrated that PYC2 expression in CHO cells can efficiently alter their metabolism and induce a desired shift towards lactate consumption that was shown to occur consistently in all culture modes and commercial media tested. Given the marked reduction in waste metabolite formation, the results presented here demonstrate that PYC2 cells are well suitable for fed-batch process employing concentrated feed solutions. Indeed, a simple feeding strategy based on glucose consumption was able to substantially increase culture growth and longevity and lead to a significant improvement of the volumetric titer. Since the cell specific productivity of PYC2 clones was marginally reduced compared to the parental cells, this enhancement was attributable to the significant increase in maximum cell density. Further improvement could potentially be achieved by screening key culture supplements or process conditions leading to increased culture longevity and/or stimulating the cell specific productivity (e.g. mild hypothermia conditions).

Acknowledgements

The authors would like to thank the following contributors: Louis Bisson for performing protein A analysis and Prof. Mario Jolicoeur (Canadian Research Chair on the development of metabolic engineering tools) for granting access to the bioreactors employed in this study. This work was supported in part by the Natural Sciences and Engineering Research Council of Canada (NSERC) MabNet Strategic Network. This is NRC publication #XXXXX.

References

Agrawal, V., Slivac, I., Perret, S., Bisson, L., St-Laurent, G., Murad, Y., Zhang, J., and Durocher, Y. (2012). Stable expression of chimeric heavy chain antibodies in CHO cells. Methods in Molecular Biology 911, 287-303.

Altamirano, C., Illanes, A., Becerra, S., Cairo, J.J., and Godia, F. (2006). Considerations on the lactate consumption by CHO cells in the presence of galactose. J Biotechnol 125, 547556.

Altamirano, C., Paredes, C., Cairo, J., and Godia, F. (2000). Improvement of CHO cell culture medium formulation: simultaneous substitution of glucose and glutamine. Biotechnology progress 16, 69-75.

Altamirano, C., Paredes, C., Illanes, A., Cairo, J.J., and Godia, F. (2004). Strategies for fed-batch cultivation of t-PA producing CHO cells: substitution of glucose and glutamine and rational design of culture medium. J Biotechnol 110, 171-179.

Andersen, D., and Goochee, C. (1995). The Effect of Ammonia on the 0-Linked G lycosylat io n of Granulocyte Colony-Stimulating Factor Produced by Chinese Hamster Ovary Cells. Biotechnology and bioengineering 47(1), 96-105.

Bell, A., Wang, Z.J., Arbabi-Ghahroudi, M., Chang, T.a., Durocher, Y., Trojahn, U., Baardsnes, J., Jaramillo, M.L., Li, S., Baral, T.N., et al. (2010). Differential tumor-targeting abilities of three single-domain antibody formats. Cancer letters 289, 81-90.

Bibila, T., and Robinson, D. (1995). In pursuit of the optimal fed-batch process for monoclonal antibody production. Biotechnolgy Progress 11, 1-13.

Borys, M., Linzer, D., and Papoutsakis, E. (1994). Ammonia affects the glycosylation patterns of recombinant mouse placental lactogen-I by chinese hamster ovary cells in a pH-dependent manner. Biotechnology and bioengineering 43, 505-514.

Chee Furng Wong, D., Tin Kam Wong, K., Tang Goh, L., Kiat Heng, C., and Gek Sim Yap, M. (2005). Impact of dynamic online fed-batch strategies on metabolism, productivity and N-glycosylation quality in CHO cell cultures. Biotechnology and bioengineering 89, 164-177.

Chen, K., Liu, Q., Xie, L., Sharp, P.a., and Wang, D.I. (2001). Engineering of a mammalian cell line for reduction of lactate formation and high monoclonal antibody production. Biotechnology and bioengineering 72, 55-61.

Chen, P., and Harcum, S.W. (2006). Effects of elevated ammonium on glycosylation gene expression in CHO cells. Metabolic engineering 8, 123-132.

Chong, W.P.K., Reddy, S.G., Yusufi, F.N.K., Lee, D.-Y., Wong, N.S.C., Heng, C.K., Yap, M.G.S., and Ho, Y.S. (2010). Metabolomics-driven approach for the improvement of Chinese hamster ovary cell growth: overexpression of malate dehydrogenase II. Journal of biotechnology 147, 116-121.

Cruz, H.J., CM Freitas, C.M., Alves, P.M., Moreira, J.L., and Carrondo, M.J.T. (2000). Effects of ammonia and lactate on growth, metabolism, and productivity of BHK cells. Enzyme and microbial technology 27, 43-52.

Dorai, H., Kyung, Y.S., Ellis, D., Kinney, C., Lin, C., Jan, D., Moore, G., and Betenbaugh, M.J. (2009). Expression of anti-apoptosis genes alters lactate metabolism of Chinese Hamster Ovary cells in culture. Biotechnology and bioengineering 103, 592-608.

Elias, C.B., Carpentier, E., Durocher, Y., Bisson, L., Wagner, R., and Kamen, A. (2003). Improving glucose and glutamine metabolism of human HEK 293 and Trichoplusia ni insect cells engineered to express a cytosolic pyruvate carboxylase enzyme. Biotechnology progress 19, 90-97.

Fogolin, M.B., Wagner, R., Etcheverrigaray, M., and Kratje, R. (2004). Impact of temperature reduction and expression of yeast pyruvate carboxylase on hGM-CSF-producing CHO cells. Journal of biotechnology 109, 179-191.

Gawlitzek, M., Ryll, T., Lofgren, J., and Sliwkowski, M. (2000). Ammonium alters N-glycan structures of recombinant TNFR-IgG: degradative versus biosynthetic mechanisms. Biotechnology and bioengineering 68, 637-646.

Genzel, Y., Ritter, J., Konig, S., Alt, R., and Reichl, U. (2005). Substitution of glutamine by pyruvate to reduce ammonia formation and growth inhibition of mammalian cells. Biotechnology progress 21, 58-69.

Hassel, l.T., and Butler, M. (1990). Adaptation to non-ammoniagenic medium and selective substrate feeding lead to enhanced yields in animal cell cultures. Journal of cell science 96, 501-508.

Henry, O., and Durocher, Y. (2011). Enhanced glycoprotein production in HEK-293 cells expressing pyruvate carboxylase. Metabolic engineering 13, 499-507.

Irani, N., Beccaria, A.J., and Wagner, R. (2002). Expression of recombinant cytoplasmic yeast pyruvate carboxylase for the improvement of the production of human erythropoietin by recombinant BHK-21 cells. Journal of biotechnology 93, 269-282.

Irani, N., Wirth, M., van Den Heuvel, J., and Wagner, R. (1999). Improvement of the primary metabolism of cell cultures by introducing a new cytoplasmic pyruvate carboxylase reaction. Biotechnology and bioengineering 66, 238-246.

Jeong, D.-w., Cho, I.T., Kim, T.S., Bae, G.W., Kim, I.-H., and Kim, I.Y. (2006). Effects of lactate dehydrogenase suppression and glycerol-3-phosphate dehydrogenase overexpression on cellular metabolism. Molecular and cellular biochemistry 284, 1-8.

Kim, S.H., and Lee, G.M. (2007a). Down-regulation of lactate dehydrogenase-A by siRNAs for reduced lactic acid formation of Chinese hamster ovary cells producing thrombopoietin. Applied microbiology and biotechnology 74, 152-159.

Kim, S.H., and Lee, G.M. (2007b). Functional expression of human pyruvate carboxylase for reduced lactic acid formation of Chinese hamster ovary cells (DG44). Applied microbiology and biotechnology 76, 659-665.

Kishishita, S., Katayama, S., Kodaira, K., Takagi, Y., Matsuda, H., Okamoto, H., Takuma, S., Hirashima, C., and Aoyagi, H. (2015). Optimization of chemically defined feed media for monoclonal antibody production in Chinese hamster ovary cells. Journal of bioscience and bioengineering.

Kurano, N., Leist, C., Messi, F., Kurano, S., and Fiechter, a. (1990). Growth behavior of Chinese hamster ovary cells in a compact loop bioreactor. 2. Effects of medium components and waste products. Journal of biotechnology 15, 113-128.

Lao, M.-S., and Toth, D. (1997). Effects of Ammonium and Lactate on Growth and Metabolism of a Recombinant Chinese Hamster Ovary Cell Culture. Biotechnology Progress 13, 688-691.

Le, H., Kabbur, S., Pollastrini, L., Sun, Z., Mills, K., Johnson, K., Karypis, G., and Hu, W.S. (2012). Multivariate analysis of cell culture bioprocess data--lactate consumption as process indicator. J Biotechnol 162, 210-223.

Liu, B., Spearman, M., Doering, J., Lattova, E., Perreault, H., and Butler, M. (2014). The availability of glucose to CHO cells affects the intracellular lipid-linked oligosaccharide distribution, site occupancy and the N-glycosylation profile of a monoclonal antibody. Journal of Biotechnology 170, 17-27.

Luo, J., Vijayasankaran, N., Autsen, J., Santuray, R., Hudson, T., Amanullah, A., and Li, F. (2012). Comparative metabolite analysis to understand lactate metabolism shift in Chinese hamster ovary cell culture process. Biotechnology and bioengineering 109, 146-156.

Ma, N., Ellet, J., Okediadi, C., Hermes, P., McCormick, E., and Casnocha, S. (2009). A single nutrient feed supports both chemically defined NS0 and CHO fed-batch processes: Improved productivity and lactate metabolism. Biotechnology progress 25, 1353-1363.

Maranga, L., and Goochee, C.F. (2006). Metabolism of PER.C6 cells cultivated under fed-batch conditions at low glucose and glutamine levels. Biotechnol Bioeng 94, 139-150.

Martinez, V.S., Dietmair, S., Quek, L.-E., Hodson, M.P., Gray, P., and Nielsen, L.K. (2013). Flux balance analysis of CHO cells before and after a metabolic switch from lactate production to consumption. Biotechnology and bioengineering 110, 660-666.

Mulukutla, B.C., Gramer, M., and Hu, W.-S. (2012). On metabolic shift to lactate consumption in fed-batch culture of mammalian cells. Metabolic engineering 14, 138-149.

Omasa, T., Higashiyama, K., Shioya, S., and Suga, K. (1991). Effects of Lactate Concentration on Hybridoma Culture in Lactate-Controlled Fed-Batch Operation. Biotechnology and bioengineering 39, 556-564.

Omasa, T., Ishimoto, M., Higashiyama, K.-i., Shioya, S., and Suga, K.-i. (1992). The enhancement of specific antibody production rate in glucose- and glutamine-controlled fed-batch culture. Cytotechnology 8, 75-84.

Ozturk, S., Riley, M., and Palsson, B. (1992). Effects of ammonia and lactate on hybridoma growth, metabolism, and antibody production. Biotechnology and bioengineering 39, 418431.

Ozturk, S., Thrift, J., Blackie, J., and Naveh, D. (1997). Real-time monitoring and control of glucose and lactate concentrations in a mammalian cell perfusion reactor. Biotechnology and bioengineering 53, 372-378.

Paredes, C., Prats, E., Cairo, J., Azorin, F., Cornudella, L., and F, G. (1999). Modification of glucose and glutamine metabolism in hybridoma cells through metabolic engineering. Cytotechnology 30, 85-93.

Pascoe, DE., Arnott, D., Papoutsakis, E.T., Miller, W.M., and Andersen, D C. (2007). Proteome analysis of antibody-producing CHO cell lines with different metabolic profiles. Biotechnol Bioeng 98, 391-410.

Vallée, C., Durocher, Y., and Henry, O. (2013). Exploiting the metabolism of PYC expressing HEK293 cells in fed-batch cultures. Journal of biotechnology 169C, 63-70.

Wilkens, C.A., and Gerdtzen, Z.P. (2015). Comparative metabolic analysis of CHO cell clones obtained through cell engineering, for IgG productivity, growth and cell longevity. PloS one 10, e0119053.

Wlaschin, K.F., and Hu, W.-S. (2006). Fedbatch Culture and Dynamic Nutrient Feeding. 101, 43-74.

Wurm, F.M. (2004). Production of recombinant protein therapeutics in cultivated mammalian cells. Nat Biotechnol 22, 1393-1398.

Xie, L., Nyberg, G., Gu, X., Li, H., Mollborn, F., and Wang, D.I. (1997). Gamma-interferon production and quality in stoichiometric fed-batch cultures of Chinese hamster ovary (CHO) cells under serum-free conditions. Biotechnology and bioengineering 56, 577-582.

Yang, M., and Butler, M. (2002). Effects of ammonia and glucosamine on the heterogeneity of erythropoietin glycoforms. Biotechnology progress 18, 129-138.

Yuk, I.H., Russell, S., Tang, Y., Hsu, W.-t., Mauger, J.B., Aulakh, R.P.S., Jun, L., Gawlitzek, M., and Joly, J.C. (2014). Effects of Copper on CHO Cells : Cellular Requirements and Product Quality Considerations. Biotechnology progress, 1-13.

Zagari, F., Jordan, M., Stettler, M., Broly, H., and Wurm, F.M. (2013a). Lactate metabolism shift in CHO cell culture: the role of mitochondrial oxidative activity. New biotechnology 30, 238-245.

Zagari, F., Stettler, M., Baldi, L., Broly, H., Wurm, F., and Jordan, M. (2013b). High expression of aspartate glutamate carrier aralar1 favors lactate consumption in CHO cell culture. Pharmaceicial Bioprocessing 1, 19-27.

Zhang, F., Sun, X., Yi, X., and Zhang, Y. (2006). Metabolic characteristics of recombinant Chinese hamster ovary cells expressing glutamine synthetase in presence and absence of glutamine. Cytotechnology 51, 21-28.

Zhou, M., Crawford, Y., Ng, D., Tung, J., Pynn, A.F.J., Meier, A., Yuk, I.H., Vijayasankaran, N., Leach, K., Joly, J., et al. (2011). Decreasing lactate level and increasing antibody production in Chinese Hamster Ovary cells (CHO) by reducing the expression of lactate dehydrogenase and pyruvate dehydrogenase kinases. Journal of biotechnology 153, 27-34.

Zhou, W., Rehm, J., and Hu, W. (1995). High viable cell concentration fed-batch cultures of hybridoma cells through on-line nutrient feeding. Biotechnology and bioengineering 46, 579587.

Figure Legends

Figure 1: Generation of a stable cell line expressing PYC2. (A) Example of western blot analysis of CHO cells transfected with the PYC2 gene. PYC2 and PC expression levels are detected using HRP-conjugated streptavidin. Lane 1: parental cell line (untransfected), lanes: 2, 4, 7, 24: PYC2 positive clones. Lanes: 3, 5, 6-23, 25: PYC2 negative clones. (B) Western blot of the 11 PYC2 positives clones chosen for batch cultivation. (C) Western blot of the 3 PYC2 clones exhibiting overexpression of PYC2 after 2 month in culture (D) Western blot analysis of PYC2 expression in the cytosolic fraction. Lanes 1-3: PYC2 positive clones. Lane 4: parental cell line (untransfected).

Figure 2: (A) Lactate and (B) viable cell concentration profiles for PYC2 clones compared to PYC2 negatives clones and the parental cell line cultivated in shake-flask cultures.

Figure 3: (A) Final lactate and (B) ammonia concentrations for PYC2 and parental cell lines cultivated in 4 different media in shake-flask cultures. Error bars depict standard deviations of the measurements. ND: medium not tested for clones 1 and 2.

£ 20 flj

20 18 E 16

¿14 ■¡12 1« I 8

I(PYC2) Clone 1 (PYC2) Clone 7 :(PYC2) Clone 11 i Parental clone

Power CHO Freestyle 17 Forti CHO CD DG44 BalanCD CD 2

Hul l - I■ I

Power CHO Freestyle 17 Forti CHO CD DG44 BalanCD

Figure 4: Evolution of pH for the parental cell line and 3 PYC2 positive clones cultivated in 6 well plates equipped with pH sensors (HydroDish®). The 3 pH peaks indicated by arrows correspond to sampling of the cultures.

—(PYC2) Clone 1 (PYC2) Clone 7 7.6 1 -(PYC2) Clone 11

■Parental clone

6.4 -I-1-1-1-1

0 2 4 6 8

Time (d)

Figure 5: (A) Cell viability, (B) cell density, (C) glucose, (D) glutamine, (E) lactate and (F) ammonia concentration profiles during batch and fed-batch cultivations of parental and PYC2-expressing cells in shake-flask cultures. Error bars depict standard deviations of the measurements.

Figure 6: (A) Cell viability, (B) cell density, (C) glucose, (D) glutamine (E) lactate and (F) ammonia concentration profiles fed-batch cultivations of parental and PYC2 expressing clones in bioreactor cultures. Error bars depict standard deviations of the measurements.

o 2 4 a e 10 12 14 ie 18 o 2 4 s e 10 12 14 is is

Time (d) Time (d)

Figure 7: Comparison of maximum antibody titers between parental and PYC2-expressing cells cultivated in shake flasks and bioreactors. Error bars depict standard deviations of the measurements.

Batch Fed-batch Fed-batch

Shake flasks Bioreactor

Table 1. Cell specific consumption/production rates in shake flask cultures

Batch Fed-batch

Exponential phase PYC2 Parental PYC2 Parental

q glucose (pmol/cell.d) 1.00 ± 0.17 1.24 ± 0.21 1.18 ± 0.19 1.65 ± 0.30

q glutamine (pmol/cell.d) 0.35 ± 0.06 0.37 ± 0.07 0.47 ± 0.08 0.53 ± 0.09

q lactate (pmol/cell.d) 0.46 ± 0.06 0.98 ± 0.18 0.54 ± 0.09 1.18 ± 0.21

q ammonia (pmol/cell.d) 0.57 ± 0.10 0.86 ± 0.18 0.93 ± 0.20 1.31 ± 0.28

lactate/glucose ratio 0.46 0.79 0.46 0.72

ammonia/glutamine ratio 1.63 2.32 1.98 2.47

Stationary phase

q glucose (pmol/cell.d) 0.69 ± 0.10 1.19 ± 0.21 1.42 ± 0.24 1.73 ± 0.29

q glutamine (pmol/cell.d) --- --- 0.27 ± 0.05 0.31 ± 0.05

q lactate (pmol/cell.d) -0.36 ± 0.06 -0.05 ± 0.01 -0.15 ± 0.03 -0.03± 0.01

q ammonia (pmol/cell.d) --- --- 0.41 ± 0.09 0.36 ± 0.08

lactate/glucose ratio -0.52 -0.04 -0.11 -0.02

ammonia/glutamine ratio --- --- 1.52 1.16

Table 2. Cell specific consumption/production rates in bioreactor cultures

Exponential phase Stationary phase

PYC2 Parental PYC2 Parental

q glucose (pmol/cell.d) 2.46 ± 0.42 2.49 ± 0.44 1.57 ± 0.27 2.06 ± 0.33

q glutamine (pmol/cell.d) 0.75 ± 0.12 1.18 ± 0.20 0.44 ± 0.08 0.56 ± 0.09

q lactate (pmol/cell.d) 0.43 ± 0.07 2.34 ± 0.41 -0.17 ± 0.03 0.24 ± 0.04

q ammonia (pmol/cell.d) 0.31 ± 0.05 1.11 ± 0.18 0.19 ± 0.03 0.24 ± 0.02

lactate/glucose ratio 0.17 0.79 -0.11 0.11

ammonia/glutamine ratio 0.41 0.94 0.43 0.82