Appl Microbiol Biotechnol DOI 10.1007/s00253-015-6617-y
BIOTECHNOLOGICAL PRODUCTS AND PROCESS ENGINEERING
Elucidating the effects of arginine and lysine on a monoclonal antibody C-terminal lysine variation in CHO cell cultures Xintao Zhang 1 & Hongping Tang 1 & Ya-Ting Sun 1 & Xuping Liu 1 & Wen-Song Tan 1 & Li Fan 1
Received: 16 March 2015 / Revised: 11 April 2015 / Accepted: 15 April 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract C-terminal lysine variants are commonly observed in monoclonal antibodies (mAbs) and found sensitive to process conditions, especially specific components in culture medium. The potential roles of media arginine (Arg) and lysine (Lys) in mAb heavy chain C-terminal lysine processing were investigated by monitoring the lysine variant levels under various Arg and Lys concentrations. Both Arg and Lys were found to significantly affect lysine variant level. Specifically, lysine variant level increased from 18.7 to 31.8 % when Arg and Lys concentrations were increased from 2 to 10 mM. Since heterogeneity of C-terminal lysine residues is due to the varying degree of proteolysis by basic carboxypeptidases (Cps), enzyme (basic Cps) level, pH conditions, and product (Arg and Lys) inhibition, which potentially affect the enzymatic reaction, were investigated under various Arg and Lys conditions. Enzyme level and pH conditions were found not to account for the different lysine variant levels, which was evident from the minimal variation in transcription level and intracellular pH. On the other hand, product inhibition effect of Arg and Lys on basic Cps was evident from the notable intracellular and extracellular Arg and Lys concentrations comparable with Ki values (inhibition constant) of basic Cps and further confirmed by cell-free assays. Additionally, a kinetic study of lysine variant level during the cell culture
Xintao Zhang and Hongping Tang contributed equally to this work. * Wen-Song Tan [email protected] * Li Fan [email protected] 1
The State Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, Shanghai 200237, China
process enabled further characterization of the C-terminal lysine processing. Keywords Chinese hamster ovary cells . Monoclonal antibody . Lysine variant . Arginine . Lysine . Carboxypeptidase
Introduction Therapeutic monoclonal antibodies (mAbs) derived from mammalian cell culture typically exhibit product heterogeneity due to multiple posttranslational modifications (PTMs) or chemical degradation, which potentially affects the product efficacy and safety (Liu et al. 2008). C-terminal lysine variant is one of the most commonly detected product heterogeneity during mAb production process. Generally, lysine residues at the heavy chain C-terminus are removed in recombinant IgGs during cell culture process (Harris 1995), while incomplete cleavage of the lysine residues leads to the presence of lysine variants and further results in charge heterogeneity (Dorai and Ganguly 2014). Although the modification of C-terminal lysine has no substantial influence on antibody structure, stability, and functions (Ponniah et al. 2014), the characterization of lysine variant level is essential to ensure the consistency of product quality. Moreover, understanding the biological roles of process parameters in Cterminal lysine processing is critical for mAb process development and manufacturing (Schenerman et al. 2004). C-terminal lysine cleavage is an enzymatic process catalyzed by intracellular and extracellular basic carboxypeptidases (Cps), including CpB, CpM, and CpH found in Chinese hamster ovary (CHO) cells (Luo et al. 2012). The activity of basic Cps is regulated by various factors, such as temperature, pH, cofactors, activators, and inhibitors (Greene et al. 1992;
Appl Microbiol Biotechnol
Wolff et al. 1962). Recently, process parameters were found to be responsible for the heterogeneity of C-terminal lysine residues by altering the activity and/or expression of basic Cps during the cell culture process (Dick et al. 2008). Luo et al. (2012) reported that the lysine variant level decreased slightly with increased culture temperature and prolonged culture duration. Besides, copper and zinc in the culture medium were also observed to have opposite effects on C-terminal lysine processing in CHO cell cultures (Luo et al. 2012). However, the potential effect of copper has been challenged by a recent study, in which no noticeable relationship was found between copper concentration and lysine variant level (Yuk et al. 2015), suggesting that process parameters may play distinct roles in C-terminal lysine processing with different products or cell lines. In addition to these reported process parameters, we found that arginine (Arg) and lysine (Lys), two basic amino acids in the culture medium, showed significant impact on lysine variant level in CHO cells during the development of chemically defined medium. Amino acids have been found to be associated with product quality, including glycosylation patterns (Aghamohseni et al. 2014; Chen and Harcum 2005; Crowell et al. 2007; Fan et al. 2015), sequence variants (Yu et al. 2009), aggregates, and fragments (Jing et al. 2012; Yang and Butler 2000). However, the effect of amino acid on mAb charge variants has not been investigated. Especially, how lysine variant is mediated by specific amino acids remains unclear. In this study, we demonstrated for the first time that mAb C-terminal lysine variant level increased with higher Arg and Lys concentrations in the culture medium. Three aspects regarding the catalytic properties of basic Cps, including enzyme level, pH conditions, and product inhibition effect, were then investigated under various Arg and Lys conditions through measurement of basic Cps transcription level, intracellular pH (pHi), and amino acid concentration, respectively. Of those investigated, product inhibition effect of Arg and Lys on basic Cps was considered to be the most significant cause for the increased lysine variant levels. A series of cell-free assays further confirmed this supposition and provided kinetic interpretation of the C-terminal lysine processing accompanied by different degrees of product inhibition. These observation and elucidation of the impact of specific amino acids on lysine variant level enable us to better understand and control product heterogeneity in cell culture process.
mixture of DMEM/Ham’s F12 (1:1, v/v) supplemented with vitamins, nucleic acids, sodium selenite, ferric citrate, and F68. Seed cells were cultivated at 37 °C in 500-mL shake flasks (Corning, NY, USA) with an agitation rate of 120 rpm in a 5 % CO2 incubator (Thermo Fisher, OH, USA). Cells were passaged every 2 days to maintain the viable cell density within a range of 0.5–2.0×106 cells/mL. Unless otherwise stated, materials used in this study were purchased from Sigma-Aldrich.
Materials and methods
Table 1 Arg and Lys concentrations of the four batch culture conditions
Cell line and cell maintenance A CHO cell line producing a chimeric anti-CD20 mAb was used in this study (Sun et al. 2013). Cells were maintained in a proprietary chemically defined medium derived from a
Batch culture Batch cultures were performed at 37 °C in 1000-mL shake flasks with working volume of 400 mL. Exponentially growing seed cells were harvested by centrifugation (1000 rpm for 5 min) and inoculated in fresh medium (without Arg and Lys) at a viable cell density of 1×106 cells/mL. Concentrated Arg and Lys solutions were immediately added to achieve various (LL, LH, HL, HH) Arg and Lys concentrations as described in Table 1. Sampling was performed daily to determine viable cell density, and the supernatant was kept at −80 °C for further analysis. Routine analytical methods Cell density and viability were measured using trypan blue dye exclusion method with a hemocytometer. Cell diameter was measured by a Countess Automated Cell Counter (Invitrogen, OR, USA). The mAb concentration was determined by protein A HPLC assay. Weak cation exchange chromatography (WCX) analysis WCX is widely used for detection and quantification of mAb variants, especially for lysine variants (Vlasak and Ionescu 2008). Since mAb with CpB treatment can entirely remove C-terminal lysine residues, the level of lysine variant can be quantified by comparing the mAb basic variants obtained from WCX with or without CpB treatment. Briefly, mAb purified by protein A affinity chromatography (GE healthcare, Uppsala, Sweden) was pretreated with CpB at a ratio of 1:100 (enzyme to mAb, w/w) to entirely remove the C-terminal lysine residues. Samples with or without CpB treatment were then both analyzed by WCX. WCX was performed on an
Condition
Arg (mM)
Lys (mM)
LL LH HL HH
2 2 10 10
2 10 2 10
Appl Microbiol Biotechnol
Agilent 1260 HPLC system using a ProPac WCX-10 column, 4.0 ×250 mm (Dionex, CA, USA). Mobile phase A was 10 mM Na 2 HPO 4 , pH = 7.5, and mobile phase B was 100 mM NaCl in 10 mM Na2HPO4, pH=7.5. A linear gradient elution from 30 to 75 % B in 20 min was used to separate mAb charge variant with detection at 280 nm. The flow rate was 1 mL/min, and the separation was performed at room temperature. Acidic variants and basic variants were defined as represented by the peaks that were eluted earlier or later than the main peak. Reverse-transcript PCR (RT-PCR) and real-time PCR (qRT-PCR) analysis RT-PCR and qRT-PCR analyses were employed to detect and quantify the transcription level of basic Cps (CpB, CpH, CpM) in CHO cells. Total cell RNA was extracted using TRIzol Reagent (Invitrogen, CA, USA) and then reverse transcribed by MMLV Reverse Transcriptase (Promega, NY, USA). RT-PCR was performed using the cDNA on a 2720 Thermal Cycler (Applied Biosystems, Singapore) with conditions as follows: 5 min at 94 °C, 30 cycles of 30 s at 94 °C, 25 s at 58 °C (53 °C for CpH), and 30 s at 72 °C, followed by 10 min at 72 °C for a final extension. GAPDH, a housekeeping gene, was used as an internal control. The primers used in this study are summarized in Table 2. The amplification products were analyzed using 1.5 % agarose gel with ethidium bromide. qRT-PCR reaction was performed and monitored using the SYBR Green PCR Master mix and the ABI Prism7700 Sequence Detection System (Applied Biosystems, CA, USA). The relative levels of basic Cps mRNAs were analyzed using the 2−ΔΔCt method with GAPDH as an internal control. Intracellular pH (pHi) measurement pHi was determined as previously described (deZengotita et al. 2002). Briefly, cells were incubated with the dye carboxyseminaphthorhodafluor-1 (carboxy-SNARF-1) (Invitrogen, CA, USA) for 30 min. The ratio of the fluorescence emission at 585 and 640 nm was then measured using a FACS Calibur flow cytometer (Becton Dickinson, CA, USA). Additionally, standards were generated by incubating cells in
Table 2
K + buffers with different pH values while nigericin (Invitrogen, CA, USA) was added to induce the equilibration of the intracellular and extracellular pH. The pHi of sample was then calculated from the fluorescence ratio of standard curve. Intracellular amino acid extraction and concentration determination Intracellular amino acids were extracted as previously described (Dietmair et al. 2010) with slight modifications. Briefly, cell fluid (5×106 cells) was quickly mixed with cold quenching solution (0.9 % (w/v) NaCl, 0 °C) (v/v, 1:4). After centrifugation (1000 g for 1 min at 0 °C), cell pellets were separated and resuspended in 1-mL ice-cold 50 % aqueous acetonitrile. The mixture was vortexed and incubated on ice for 10 min. After centrifugation (20000g for 5 min at 0 °C), supernatant was separated and dried by a SpeedVac (Thermo Savant, NY, USA). The dried extracts were resuspended in water for further analysis. Concentrations of intracellular and extracellular amino acids were quantified using an Agilent 1260 HPLC coupled with a Nova-Pak C18 column, 3.9×150 mm (Waters, MA, USA), following the precolumn derivation using AccQ·Fluor reagent kit (Waters, MA, USA). Intracellular amino acid concentrations were calculated based on cell volume by presuming that each cell was spherical with determined diameter (Hansen and Emborg 1994). Cell-free assay The batch culture under LL condition (Table 1) was designated as the cell process (CP) and performed as described. On day 6, 100-mL cell fluid (CF) from CP was centrifuged (1000 rpm for 5 min) to discard cells and sterile filtered (0.22 μm) as harvested cell fluid (HCF). The HCF was then divided into four parts and added additional Arg and Lys, i.e., three parts with additional 8 mM Arg, 8 mM Lys, 8 mM Arg and Lys, respectively, and the remaining part with no addition. The four HCFs were then reincubated under the same condition of CP for another 8 days. Samples were taken daily from the HCFs to determine the lysine variant level during the incubation.
Primers used for analysis of basic Cps in CHO cells
Gene
Forward primers (5′–3′)
Reverse primers (5′–3′)
Product length (bp)
GAPDH CpB CpH CpM
AGGTTGTCTCCTGCGACTTCA CAGCGAGAATCCACAACTCA TGTGCTCTCCGCCAATCT CCGATACCACCACCAGGAAG
GAGGTCCACCACTCTGTTGCT ATCAATGTTGACCACAGGCA AACTACTGTCATCGTCGTTCTT CCGAGTGCTGTTGATTAGTTGT
143 265 203 305
Appl Microbiol Biotechnol
For the kinetic study of C-terminal lysine processing during the cell culture process, four batch cultures with various Arg and Lys concentrations (Table 1) were performed as described above. A 50-mL CF was taken on days 1, 3, 5, 6, 7, 8 from each condition, prepared as HCF and reincubated under the same condition of CP. Samples were taken daily from the HCFs to determine the lysine variant level during the incubation.
Results Cell growth, mAb production, and lysine variant level To determine the potential roles of Arg and lysin C-terminal lysine processing during the cell culture process, batch cultures with various Arg and Lys concentrations were performed as illustrated in Table 1. Cultures were performed in triplicate. Figure 1 shows the profiles of cell growth, mAb production, and lysine variant level. No statistically significant difference of cell growth and mAb production was found in cultures with various concentrations of Arg and Lys (Fig. 1a, b). The maximum viable cell densities reached 3.5– 4.2×106 cells/mL on day 4 under the four conditions (Fig. 1a). Meanwhile, the accumulated mAb concentrations reached 211–235 mg/L during the 8-day culture (Fig. 1b). Contrary to the similar cell growth and mAb production, significant differences of lysine variant levels were observed under the four conditions on day 8 (Fig. 1c). Specifically, only 18.7 % lysine variant level was observed when both Arg and Lys concentrations were low (2 mM Arg and 2 mM Lys, defined as LL). However, when 10 mM Arg or Lys was added (LH or HL), the lysine variant level increased significantly to 28.3 or 23.6 % (p=0.003, 0.015), and even reached up to 31.8 % in the case of both high Arg and Lys concentrations (HH). These variations indicate a positive effect of Arg and Lys on mAb lysine variant level. Moreover, it can also be speculated that Lys has obviously stronger influence than Arg (p=0.014). Since heterogeneity of C-terminal lysine residues is believed to result from varying degree of proteolysis by basic Cps, enzyme level, pH conditions, and product (Arg and Lys) inhibition, which potentially affect the enzymatic reaction, were then investigated under various Arg and Lys conditions to elucidate this positive effect of Arg and Lys on mAb lysine variant level. Basic Cps detection by RT-PCR assay In this study, the presence of normal basic Cps in CHO cells (including CpB, CpH, and CpM) was first examined by RTPCR assay. The sample was taken from the batch culture under LL condition (Table 1) on day 6, and the result is shown in Fig. 2a. The lengths of the PCR products of approximately 260 bp (CpB) and 200 bp (CpH) obtained from the
Fig. 1 Profiles of a cell growth, b mAb production, and c lysine variant level during the batch cultures with various Arg and Lys concentrations. Black square LL, black circle LH, black triangle HL, black diamond HH. Note that the error bars indicate the standard deviations from three independent experiments, and p values are estimated by two-tailed Student’s t test
electropherogram were in accordance with the designed values (Table 2), thus indicating the presence of CpB and CpH in the CHO cell line we used. However, no apparent signal for CpM was obtained. The transcription levels of CpB and CpH during the batch cultures with various Arg and Lys concentrations were further investigated by qRT-PCR, and the results are shown in Fig. 2b, c. As can be seen, the relative mRNA levels (to GAPDH) of CpB and CpH were stable throughout the 8-day culture, and no significant difference existed among the four conditions, indicating that the transcription levels of CpB and CpH should not account for the different lysine variant levels.
Appl Microbiol Biotechnol
Fig. 3 Profile of pHi during the batch cultures with various Arg and Lys concentrations. Black square LL, black circle LH, black triangle HL, black diamond HH. Note that the values presented are the average from two independent experiments
pHi (no more than 0.05 unit) was believed not to significantly affect the activity of basic Cps (Bradley et al. 1996; Wolff et al. 1962). Additionally, extracellular pH (pHe) was also measured during the batch cultures, and no significant difference was observed among the four conditions (data not shown), thus indicating that the different lysine variant levels should not be attributed to pH condition.
Amino acid concentration
Fig. 2 RT-PCR and qRT-PCR analysis of basic Cps in CHO cells with GAPDH as an internal control. a Typical agarose gel electropherogram for the detection of CpB, CpH, and CpM, with a 50-bp DNA ladder. Relative b CpB and c CpH mRNA levels during the batch cultures with various Arg and Lys concentrations. The mRNA levels for each gene were normalized to LL condition on day 1. □: LL, : LH, : HL, ■: HH. Note that the error bars indicate the standard deviations from three independent experiments
Variations in intracellular pH (pHi) The pHi values during the batch cultures with various Arg and Lys concentrations are monitored and shown in Fig. 3. It was found that the pHi values decreased gradually throughout the culture processes regardless of conditions. From an initial value of 7.46, the pHi decreased to 7.40 (LL), 7.36 (LH), 7.37 (HL), and 7.35 (HH) on day 8 under the four conditions. A lower pHi value was observed throughout the culture process under higher Arg and Lys concentrations (HH) when compared with the other conditions. However, this difference in
As the catalytic products of basic Cps, Arg and Lys were reported to have the potential to inhibit the activity of basic Cps in a dose-dependent manner (Hook 1990; Wolff et al. 1962). Therefore, intracellular and extracellular amino acid concentrations were quantified to determine whether these two basic amino acids reached the inhibition concentration and consequently affected the lysine variant level. Samples under the four culture conditions were taken every 2 days, and the results are presented in Fig. 4. For an initial concentration of 2 or 10 mM in the medium, extracellular Arg concentration decreased to approximately 1 or 9 mM on day 7 (Fig. 4a), whereas intracellular concentration increased to approximately 2 or 9 mM on day 3 and then decreased to approximately 1 or 4 mM on day 7 (Fig. 4c). Similarly, for an initial concentration of 2 or 10 mM in the medium, extracellular Lys concentration decreased to approximately 0.5 or 7 mM on day 7 (Fig. 4b), whereas intracellular concentration increased to approximately 1 or 4 mM on day 3 and then decreased to approximately 0.5 or 3 mM on day 7 (Fig. 4d). Altogether, the intracellular and extracellular Arg and Lys concentrations ranged from 0.5 to 2 mM for lower initial concentration (2 mM). For higher initial concentration (10 mM), the intracellular and extracellular Arg and Lys concentrations ranged from 3 to 10 mM, which were about 4-fold higher than those for lower concentration.
Appl Microbiol Biotechnol
Fig. 4 a Extracellular Arg, b extracellular Lys, c intracellular Arg, and d intracellular Lys concentrations during the batch cultures. □: LL, : LH, : HL, ■: HH. Note that the error bars indicate the standard deviations from three independent experiments
Determined by the published works, the Ki values (inhibition constant) of CpB from porcine were 0.5 and 13 mM for Arg and Lys, respectively (Wolff et al. 1962), and the Ki values of CpH from bovine were 4.6 and 7.6 mM for Arg and Lys, respectively (Hook 1990). In the present study, the intracellular and extracellular Arg and Lys concentrations ranged from 0.5 to 10 mM during the batch cultures and were the same order of magnitude (mM) as the reported Ki values, indicating the possibility of occurrence of product inhibition effect of Arg and Lys on basic Cps. Therefore, the significant differences in intracellular and extracellular concentrations of the two basic amino acids under the four conditions, as discussed above, would lead to different extents of inhibition, thus resulting in different lysine variant levels. However, the inhibition effect of Arg and Lys on basic Cps in CHO cells should be demonstrated with direct evidence. Moreover, the C-terminal lysine processing during the culture process should be further dynamically characterized. Cell-free assay In order to confirm the inhibition effect of Arg and Lys on basic Cps in CHO cells, a cell-free assay of HCF incubation was performed with various Arg and Lys concentrations.
Lysine variant levels during the incubation with various concentrations of Arg and Lys are determined and shown in Fig. 5a. It was found that the lysine variant level decreased from 15.6 to 8.2 % in the end, at an average rate of 0.9 %/day in the absence of additional Arg and Lys, indicating the presence of extracellular basic Cps in the HCF and its activity of C-terminal lysine cleavage. However, when additional 8 mM Lys, 8 mM Arg, or 8 mM Arg and Lys were added, the lysine variant levels decreased from 15.6 to 9.7 %, 9.4 or 10.4 % in the end, at a lower rate of 0.7 %/day, 0.8 %/day, or 0.6 %/day compared with the sample with no addition, demonstrating that the activity of extracellular basic Cps was inhibited by these two basic amino acids. Furthermore, a linear regression of the logarithm of lysine variant level (%) against incubation time shows that the extracellular C-terminal lysine cleavage process is a first-order reaction with the rate coefficient (k) represented by the fitted slope (Fig. 5b). Therefore, the extracellular process is significantly inhibited by Arg and Lys because the k value decreased by 37 % with additional 8 mM Arg and Lys when compared with the sample with no addition. Moreover, a greater reduction (25 %) in k value with additional 8 mM Lys than that (21 %) with equivalent Arg also indicates that Lys is a more effective inhibitor of basic Cps in CHO cells.
Appl Microbiol Biotechnol
various Arg and Lys concentrations are also determined and shown in Fig. 6c. As under the LL condition, the k values also increased under the other conditions (LH, HL, HH) during the culture process. Moreover, higher concentrations of Arg and Lys led to a lower k value, which should be attributed to the inhibition effect of the two basic amino acids.
Discussion
Fig. 5 Time course of HCF lysine variant level during incubation with various Arg and Lys concentrations. a Profile of lysine variant level. b Plot of the lysine variant level on a log scale. Black square no additional Arg and Lys, black circle 8 mM additional Lys, black triangle 8 mM additional Arg, black diamond 8 mM additional Arg and Lys. Note that the values presented are the average from three independent experiments
C-terminal lysine processing is a complicated process that occurs both intracellularly and extracellularly. To further characterize the C-terminal lysine processing, we conducted a kinetic study of the C-terminal lysine processing during the cell culture process. As shown in Fig. 6a, lysine variant level of CP under LL condition increased from 8.9 % on day 1 to 17.7 % on day 9. Contrarily, the lysine variant levels of HCFs gradually decreased correspondingly owing to the extracellular basic Cps activity. By plotting the logarithm of lysine variant level (%) against culture time (Fig. 6b), the k values of the extracellular C-terminal lysine processing during the batch culture are estimated and illustrated in Fig. 6c. An increase in the k value from 0.052 day−1 on day 1 to 0.086 day−1 on day 8 (condition LL) indicates that the rate of extracellular Cterminal lysine cleavage would increase during the culture process since it positively correlates with the k value (Fig. 6c) and lysine variant level (Fig. 6a). Therefore, considering that the measured lysine variant level in the CP increased during the culture process (Fig. 6a) regardless of the increased extracellular lysine cleavage rate (Fig. 6c), the lysine variant level of mAb newly secreted in the CP would also increase probably because of a reduction in the intracellular cleavage rate. Additionally, the k values of the batch cultures with
Arg and Lys, two basic amino acids, have been well characterized by their noticeable roles in cell growth and mAb production during the cell culture process (Carrillo-Cocom et al. 2014; Gonzalez-Leal et al. 2011). They were also found to stabilize mAbs during the purification and formulation process (Falconer et al. 2011). However, their effect on mAb quality, especially charge variants, has been rarely discovered. In this study, to obtain a detailed understanding of their potential roles in the C-terminal lysine processing during cell culture process, a typical antibody-producing CHO cell line was employed and cultivated with various Arg and Lys concentrations. Contrary to the unaffected cell growth and mAb production, of particular interest finding was that the lysine variant levels varied significantly with various Arg and Lys concentrations. A higher level of lysine variant was detected in the culture with higher concentrations of Arg and Lys, a phenomenon which is the first time to report to our best knowledge. As an enzymatic process, C-terminal lysine cleavage is potentially affected by various factors, such as temperature, pH, substrate and enzyme levels, and inhibition and activation effects. Accordingly, possible causes for the elevated lysine variant levels under higher Arg and Lys concentration conditions were further investigated. In this study, RT-PCR assay was performed to detect the basic Cps in CHO cells. Previous Western blot results and CHO genome database data indicated the possible presence of CpB, CpM, and CpH in CHO cells (Luo et al. 2012). Our RT-PCR results demonstrated the transcription of both CpB and CpH, confirming the presence of these two basic Cps in CHO cells. However, no apparent signal for CpM was found, which might be attributed to the low expression level in the CHO cell line we used. Furthermore, the transcription levels of CpB and CpH during the batch cultures were determined and showed no significant difference among the cultures with various concentrations of Arg and Lys (Fig. 2b, c). The transcription process is typically regarded as the dominant factor controlling protein expression (Jiang et al. 2006), and RT-PCR is widely used for relative and quantitative measurement of gene expression (Wong et al. 2010a; Wong et al. 2010b). The present result revealed that high concentrations of Arg and Lys in culture medium had no obvious influence on the transcription of CpB and CpH in CHO cells. However, the actual
Appl Microbiol Biotechnol
Fig. 6 Kinetic study of C-terminal lysine processing during the batch cultures with various Arg and Lys concentrations. a Time course of CP and HCF lysine variant level during incubation under LL condition. b Plot of the lysine variant level of HCF during incubation on a log scale. c k values of extracellular C-terminal lysine cleavage during the batch cultures with various Arg and Lys concentrations. a, b Black diamond
CP, black square HCF on day 1, black circle HCF on day 3, black triangle HCF on day 5, white square HCF on day 6, white circle HCF on day 7, white triangle HCF on day 8. c black square LL, black circle LH, black triangle HL, black diamond HH. Note that the values presented are the average from three independent experiments
activity of CpB and CpH might be affected by other factors that correlate more with Arg and Lys. The potential difference in pHi and pHe caused by various levels of Arg and Lys was another concern which might influence the C-terminal lysine processing by affecting the activity of basic Cps. Exist as zwitterions at physiological pH, amino acids act as buffers of pHi (Chen and Harcum 2005; Edwards et al. 1998). High levels of Arg and Lys, two basic amino acids with pI values of 10.76 and 9.74, respectively, had the potential to raise the pHi and therefore influence the activity of basic Cps. Surprisingly, a slight decrease (no more than 0.05 unit) in pHi was observed when cultivated with higher levels of Arg and Lys (Fig. 3). This decrease in pHi might be caused by the increased accumulation of acidic metabolite (e.g., lactic acid), which might outweigh the opposite effect caused by the higher levels of Arg and Lys (Jeong et al. 2006). Nevertheless, such a difference in pHi was believed not to significantly affect the activity of basic Cps (Bradley et al. 1996; Wolff et al. 1962).
On the other hand, the activity of basic Cps might be competitively inhibited by Arg and Lys depending on the inhibitor concentration. Actually, product inhibition effects of CpB and CpH are ubiquitous and play important roles in biological active peptide synthesis (Hook and LaGamma 1987; Wolff et al. 1962). Arg and Lys, the catalytic products of basic Cps, have been previously found to effectively inhibit the activity of basic Cps at the mM order of magnitude (Hook and LaGamma 1987; Wolff et al. 1962). In the present study, remarkable intracellular and extracellular Arg and Lys concentrations were observed comparable with Ki values of basic Cps. Although the Ki values were from bovine and porcine, they were of the same order as those from other species due to the high degree of homology of basic Cps from different species (Bradley et al. 1996; Hook 1988). Therefore, the present results suggested the possible product inhibition of basic Cps by Arg and Lys during the cell culture process. Accordingly, the different lysine variant levels (Fig. 1c) could be attributed to the different inhibition effects caused by various
Appl Microbiol Biotechnol
concentrations of Arg and Lys. Furthermore, in this study, Lys showed as a more effective inhibitor than Arg in inhibiting the C-terminal lysine cleavage (Fig. 1c). However, according to the reported study, the Ki values of the two Cps from multiple sources are generally smaller for Arg than those for Lys, indicating that Arg is a more effective inhibitor (Hook and LaGamma 1987; Wolff et al. 1962). These inconsistent results indicate the possible difference in the mechanism of product inhibition within different species, which is worth further investigation and verification. Moreover, a series of cell-free assays of HCF incubation were performed to confirm the product inhibition effect of Arg and Lys in CHO cells and to further characterize the Cterminal lysine processing during the culture process. Cellfree assay seems to be a powerful approach to simplify complicated PTMs, including C-terminal lysine processing, which occur both intracellularly and extracellularly (Dick et al. 2008; Luo et al. 2012). The significant inhibition effect of Arg and Lys on the extracellular basic Cps was verified using the kinetic analysis. As expected, the kinetic analysis also revealed that the extracellular C-terminal lysine processing is a firstorder reaction, which is consistent with the lysine processing of IgG2 in vivo (Cai et al. 2011). Moreover, the intracellular C-terminal lysine processing might also be a first-order reaction on the basis of the mechanism of the enzymatic reaction. Additionally, a kinetic study of lysine variant level during the cell culture process enabled thorough understanding of the Cterminal lysine processing. It was found that the extracellular cleavage rate increased while the intracellular cleavage rate decreased during the culture process, which might be attributed to the distinct intracellular and extracellular variations of various factors, such as enzyme, cofactor, and inhibitor levels, in addition to Arg and Lys concentrations. Although the inhibition effects of Arg and Lys on the intracellular basic Cps remain unconfirmed, the inhibition effect on the extracellular C-terminal lysine processing clearly plays an important role in modulating the final lysine variant level. Besides the mAb used in this study, we have also investigated the product inhibition effect on other IgGs, and it seems to be an un-neglectable issue during typical CHO cell cultures. More importantly, much attention should be paid to these two basic amino acids because of their influence on both productivity and quality of mAb during culture process. Maintaining the concentrations of Arg and Lys at a low level will be an effective approach for reducing the lysine variant level. However, simply reducing the concentration in medium may cause growth arrest and subsequent low mAb concentration, since Arg and Lys are two essential amino acids in cell culture process with demonstrated positive impact on cell growth and mAb production (Carrillo-Cocom et al. 2014; GonzalezLeal et al. 2011). Therefore, optimization of culture process, especially the medium component, under the guidance of quality by design (QbD) seems to be a better approach to
alleviate the product inhibition of Arg and Lys (Rouiller et al. 2012). Via the balance of various amino acids and other nutrients, the reduced lysine variant level can be achieved, and the consistency of the culture process can also be well controlled. In conclusion, high concentrations of Arg and Lys significantly enhanced mAb lysine variant level by product inhibition effect on basic Cps in CHO cell cultures. This observation and elucidation of the impact of specific amino acids on lysine variant level help to better understand and control product heterogeneity in mAb manufacturing.
Acknowledgments This work was supported by the National Natural Science Foundation of China (No. 21406066, 21206040), the National High Technology Research and Development Program of China (863 Program) (No. 2012AA02A303). Conflict of interest The authors have no conflict of interest.
References Aghamohseni H, Ohadi K, Spearman M, Krahn N, Moo-Young M, Scharer JM, Butler M, Budman HM (2014) Effects of nutrient levels and average culture pH on the glycosylation pattern of camelidhumanized monoclonal antibody. J Biotechnol 186:98–109 Bradley G, Naudé RJ, Muramoto K, Yamauchi F, Oelofsen W (1996) Ostrich (Struthio camelus) carboxypeptidase B: Purification, kinetic properties and characterization of the pancreatic enzyme. Int J Biochem Cell Biol 28(5):521–529 Cai B, Pan H, Flynn GC (2011) C-terminal lysine processing of human immunoglobulin G2 heavy chain in vivo. Biotechnol Bioeng 108(2):404–412 Carrillo-Cocom LM, Genel-Rey T, Araiz-Hernandez D, Lopez-Pacheco F, Lopez-Meza J, Rocha-Pizana MR, Ramirez-Medrano A, Alvarez MM (2014) Amino acid consumption in naive and recombinant CHO cell cultures: producers of a monoclonal antibody. Cytotechnology. doi:10.1007/s10616-014-9720-5 Chen P, Harcum SW (2005) Effects of amino acid additions on ammonium stressed CHO cells. J Biotechnol 117(3):277–286 Crowell CK, Grampp GE, Rogers GN, Miller J, Scheinman RI (2007) Amino acid and manganese supplementation modulates the glycosylation state of erythropoietin in a CHO culture system. Biotechnol Bioeng 96(3):538–549 deZengotita VM, Schmelzer AE, Miller WM (2002) Characterization of hybridoma cell responses to elevated pCO2 and osmolality: intracellular pH, cell size, apoptosis, and metabolism. Biotechnol Bioeng 77(4):369–380 Dick LW Jr, Qiu D, Mahon D, Adamo M, Cheng KC (2008) C-terminal lysine variants in fully human monoclonal antibodies: investigation of test methods and possible causes. Biotechnol Bioeng 100(6): 1132–1143 Dietmair S, Timmins NE, Gray PP, Nielsen LK, Krömer JO (2010) Towards quantitative metabolomics of mammalian cells: Development of a metabolite extraction protocol. Anal Biochem 404(2):155–164 Dorai H, Ganguly S (2014) Mammalian cell-produced therapeutic proteins: heterogeneity derived from protein degradation. Curr Opin Biotechnol 30:198–204
Appl Microbiol Biotechnol Edwards LJ, Williams DA, Gardner DK (1998) Intracellular pH of the mouse preimplantation embryo: amino acids act as buffers of intracellular pH. Hum Reprod 13(12):3441–3448 Falconer RJ, Chan C, Hughes K, Munro TP (2011) Stabilization of a monoclonal antibody during purification and formulation by addition of basic amino acid excipients. J Chem Technol Biotechnol 86(7):942–948 Fan Y, Jimenez Del Val I, Muller C, Wagtberg Sen J, Rasmussen SK, Kontoravdi C, Weilguny D, Andersen MR (2015) Amino acid and glucose metabolism in fed-batch CHO cell culture affects antibody production and glycosylation. Biotechnol Bioeng 112(3):521–535 Gonzalez-Leal IJ, Carrillo-Cocom LM, Ramirez-Medrano A, LopezPacheco F, Bulnes-Abundis D, Webb-Vargas Y, Alvarez MM (2011) Use of a Plackett-Burman statistical design to determine the effect of selected amino acids on monoclonal antibody production in CHO cells. Biotechnol Prog 27(6):1709–1717 Greene D, Das B, Fricker LD (1992) Regulation of carboxypeptidase E. Effect of pH, temperature and Co2+ on kinetic parameters of substrate hydrolysis. Biochem J 285:613–618 Hansen H, Emborg C (1994) Extra-and intracellular amino acid concentrations in continuous Chinese hamster ovary cell culture. Appl Microbiol Biotechnol 41(5):560–564 Harris RJ (1995) Processing of C-terminal lysine and arginine residues of proteins isolated from mammalian cell culture. J Chromatogr A 705(1):129–134 Hook V, LaGamma E (1987) Product inhibition of carboxypeptidase H. J Biol Chem 262(26):12583–12588 Hook VY (1988) Regulation of carboxypeptidase H by inhibitory and stimulatory mechanisms during neuropeptide precursor processing. Cell Mol Neurobiol 8(1):49–55 Hook VY (1990) Arginine and lysine product inhibition of bovine adrenomedullary carboxypeptidase H, a prohormone processing enzyme. Life Sci 47(13):1135–1139 Jeong DW, Cho IT, Kim TS, Bae GW, Kim IH, Kim IY (2006) Effects of lactate dehydrogenase suppression and glycerol-3-phosphate dehydrogenase overexpression on cellular metabolism. Mol Cell Biochem 284(1-2):1–8 Jiang Z, Huang Y, Sharfstein ST (2006) Regulation of recombinant monoclonal antibody production in chinese hamster ovary cells: a comparative study of gene copy number, mRNA level, and protein expression. Biotechnol Prog 22(1):313–318 Jing Y, Borys M, Nayak S, Egan S, Qian Y, Pan S-H, Li ZJ (2012) Identification of cell culture conditions to control protein aggregation of IgG fusion proteins expressed in Chinese hamster ovary cells. Process Biochem 47(1):69–75
Liu H, Gaza-Bulseco G, Faldu D, Chumsae C, Sun J (2008) Heterogeneity of monoclonal antibodies. J Pharm Sci 97(7):2426– 2447 Luo J, Zhang J, Ren D, Tsai WL, Li F, Amanullah A, Hudson T (2012) Probing of C-terminal lysine variation in a recombinant monoclonal antibody production using Chinese hamster ovary cells with chemically defined media. Biotechnol Bioeng 109(9):2306–2315 Ponniah G, Zhang H-M, Nowak C, Neill A, Gonzalez-Lopez N, Patel R, Chang G, Kita A, Andrien B, Liu H (2014) In vitro and in vivo modifications of recombinant and human IgG antibodies. mAbs 6(5):1–11 Rouiller Y, Solacroup T, Deparis V, Barbafieri M, Gleixner R, Broly H, Eon-Duval A (2012) Application of Quality by Design to the characterization of the cell culture process of an Fc-Fusion protein. Eur J Pharm Biopharm 81(2):426–437 Schenerman MA, Sunday BR, Kozlowski S, Webber K, GazzanoSantoro H, Mire-Sluis A (2004) CMC strategy forum report. BioProc Int 2(2):42–52 Sun Y-t, Zhao L, Ye Z, Fan L, Liu X-p, Tan W-S (2013) Development of a fed-batch cultivation for antibody-producing cells based on combined feeding strategy of glucose and galactose. Biochem Eng J 81:126–135 Vlasak J, Ionescu R (2008) Heterogeneity of monoclonal antibodies revealed by charge-sensitive methods. Curr Pharm Biotechnol 9(6): 468–481 Wolff EC, Schirmer E, Folk J (1962) The kinetics of carboxypeptidase B activity I. Kinetic parameters. J Biol Chem 237(10):3094–3099 Wong DC, Wong NS, Goh JS, May LM, Yap MG (2010a) Profiling of Nglycosylation gene expression in CHO cell fed-batch cultures. Biotechnol Bioeng 107(3):516–528 Wong NS, Wati L, Nissom PM, Feng HT, Lee MM, Yap MG (2010b) An investigation of intracellular glycosylation activities in CHO cells: effects of nucleotide sugar precursor feeding. Biotechnol Bioeng 107(2):321–336 Yang M, Butler M (2000) Enhanced erythropoietin heterogeneity in a CHO culture is caused by proteolytic degradation and can be eliminated by a high glutamine level. Cytotechnology 34(1-2):83–99 Yu XC, Borisov OV, Alvarez M, Michels DA, Wang YJ, Ling V (2009) Identification of codon-specific serine to asparagine mistranslation in recombinant monoclonal antibodies by high-resolution mass spectrometry. Anal Chem 81(22):9282–9290 Yuk IH, Russell S, Tang Y, Hsu WT, Mauger JB, Aulakh RP, Luo J, Gawlitzek M, Joly JC (2015) Effects of copper on CHO cells: Cellular requirements and product quality considerations. Biotechnol Prog 31(1):226–238
BIOTECHNOLOGICAL PRODUCTS AND PROCESS ENGINEERING
Elucidating the effects of arginine and lysine on a monoclonal antibody C-terminal lysine variation in CHO cell cultures Xintao Zhang 1 & Hongping Tang 1 & Ya-Ting Sun 1 & Xuping Liu 1 & Wen-Song Tan 1 & Li Fan 1
Received: 16 March 2015 / Revised: 11 April 2015 / Accepted: 15 April 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract C-terminal lysine variants are commonly observed in monoclonal antibodies (mAbs) and found sensitive to process conditions, especially specific components in culture medium. The potential roles of media arginine (Arg) and lysine (Lys) in mAb heavy chain C-terminal lysine processing were investigated by monitoring the lysine variant levels under various Arg and Lys concentrations. Both Arg and Lys were found to significantly affect lysine variant level. Specifically, lysine variant level increased from 18.7 to 31.8 % when Arg and Lys concentrations were increased from 2 to 10 mM. Since heterogeneity of C-terminal lysine residues is due to the varying degree of proteolysis by basic carboxypeptidases (Cps), enzyme (basic Cps) level, pH conditions, and product (Arg and Lys) inhibition, which potentially affect the enzymatic reaction, were investigated under various Arg and Lys conditions. Enzyme level and pH conditions were found not to account for the different lysine variant levels, which was evident from the minimal variation in transcription level and intracellular pH. On the other hand, product inhibition effect of Arg and Lys on basic Cps was evident from the notable intracellular and extracellular Arg and Lys concentrations comparable with Ki values (inhibition constant) of basic Cps and further confirmed by cell-free assays. Additionally, a kinetic study of lysine variant level during the cell culture
Xintao Zhang and Hongping Tang contributed equally to this work. * Wen-Song Tan [email protected] * Li Fan [email protected] 1
The State Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, Shanghai 200237, China
process enabled further characterization of the C-terminal lysine processing. Keywords Chinese hamster ovary cells . Monoclonal antibody . Lysine variant . Arginine . Lysine . Carboxypeptidase
Introduction Therapeutic monoclonal antibodies (mAbs) derived from mammalian cell culture typically exhibit product heterogeneity due to multiple posttranslational modifications (PTMs) or chemical degradation, which potentially affects the product efficacy and safety (Liu et al. 2008). C-terminal lysine variant is one of the most commonly detected product heterogeneity during mAb production process. Generally, lysine residues at the heavy chain C-terminus are removed in recombinant IgGs during cell culture process (Harris 1995), while incomplete cleavage of the lysine residues leads to the presence of lysine variants and further results in charge heterogeneity (Dorai and Ganguly 2014). Although the modification of C-terminal lysine has no substantial influence on antibody structure, stability, and functions (Ponniah et al. 2014), the characterization of lysine variant level is essential to ensure the consistency of product quality. Moreover, understanding the biological roles of process parameters in Cterminal lysine processing is critical for mAb process development and manufacturing (Schenerman et al. 2004). C-terminal lysine cleavage is an enzymatic process catalyzed by intracellular and extracellular basic carboxypeptidases (Cps), including CpB, CpM, and CpH found in Chinese hamster ovary (CHO) cells (Luo et al. 2012). The activity of basic Cps is regulated by various factors, such as temperature, pH, cofactors, activators, and inhibitors (Greene et al. 1992;
Appl Microbiol Biotechnol
Wolff et al. 1962). Recently, process parameters were found to be responsible for the heterogeneity of C-terminal lysine residues by altering the activity and/or expression of basic Cps during the cell culture process (Dick et al. 2008). Luo et al. (2012) reported that the lysine variant level decreased slightly with increased culture temperature and prolonged culture duration. Besides, copper and zinc in the culture medium were also observed to have opposite effects on C-terminal lysine processing in CHO cell cultures (Luo et al. 2012). However, the potential effect of copper has been challenged by a recent study, in which no noticeable relationship was found between copper concentration and lysine variant level (Yuk et al. 2015), suggesting that process parameters may play distinct roles in C-terminal lysine processing with different products or cell lines. In addition to these reported process parameters, we found that arginine (Arg) and lysine (Lys), two basic amino acids in the culture medium, showed significant impact on lysine variant level in CHO cells during the development of chemically defined medium. Amino acids have been found to be associated with product quality, including glycosylation patterns (Aghamohseni et al. 2014; Chen and Harcum 2005; Crowell et al. 2007; Fan et al. 2015), sequence variants (Yu et al. 2009), aggregates, and fragments (Jing et al. 2012; Yang and Butler 2000). However, the effect of amino acid on mAb charge variants has not been investigated. Especially, how lysine variant is mediated by specific amino acids remains unclear. In this study, we demonstrated for the first time that mAb C-terminal lysine variant level increased with higher Arg and Lys concentrations in the culture medium. Three aspects regarding the catalytic properties of basic Cps, including enzyme level, pH conditions, and product inhibition effect, were then investigated under various Arg and Lys conditions through measurement of basic Cps transcription level, intracellular pH (pHi), and amino acid concentration, respectively. Of those investigated, product inhibition effect of Arg and Lys on basic Cps was considered to be the most significant cause for the increased lysine variant levels. A series of cell-free assays further confirmed this supposition and provided kinetic interpretation of the C-terminal lysine processing accompanied by different degrees of product inhibition. These observation and elucidation of the impact of specific amino acids on lysine variant level enable us to better understand and control product heterogeneity in cell culture process.
mixture of DMEM/Ham’s F12 (1:1, v/v) supplemented with vitamins, nucleic acids, sodium selenite, ferric citrate, and F68. Seed cells were cultivated at 37 °C in 500-mL shake flasks (Corning, NY, USA) with an agitation rate of 120 rpm in a 5 % CO2 incubator (Thermo Fisher, OH, USA). Cells were passaged every 2 days to maintain the viable cell density within a range of 0.5–2.0×106 cells/mL. Unless otherwise stated, materials used in this study were purchased from Sigma-Aldrich.
Materials and methods
Table 1 Arg and Lys concentrations of the four batch culture conditions
Cell line and cell maintenance A CHO cell line producing a chimeric anti-CD20 mAb was used in this study (Sun et al. 2013). Cells were maintained in a proprietary chemically defined medium derived from a
Batch culture Batch cultures were performed at 37 °C in 1000-mL shake flasks with working volume of 400 mL. Exponentially growing seed cells were harvested by centrifugation (1000 rpm for 5 min) and inoculated in fresh medium (without Arg and Lys) at a viable cell density of 1×106 cells/mL. Concentrated Arg and Lys solutions were immediately added to achieve various (LL, LH, HL, HH) Arg and Lys concentrations as described in Table 1. Sampling was performed daily to determine viable cell density, and the supernatant was kept at −80 °C for further analysis. Routine analytical methods Cell density and viability were measured using trypan blue dye exclusion method with a hemocytometer. Cell diameter was measured by a Countess Automated Cell Counter (Invitrogen, OR, USA). The mAb concentration was determined by protein A HPLC assay. Weak cation exchange chromatography (WCX) analysis WCX is widely used for detection and quantification of mAb variants, especially for lysine variants (Vlasak and Ionescu 2008). Since mAb with CpB treatment can entirely remove C-terminal lysine residues, the level of lysine variant can be quantified by comparing the mAb basic variants obtained from WCX with or without CpB treatment. Briefly, mAb purified by protein A affinity chromatography (GE healthcare, Uppsala, Sweden) was pretreated with CpB at a ratio of 1:100 (enzyme to mAb, w/w) to entirely remove the C-terminal lysine residues. Samples with or without CpB treatment were then both analyzed by WCX. WCX was performed on an
Condition
Arg (mM)
Lys (mM)
LL LH HL HH
2 2 10 10
2 10 2 10
Appl Microbiol Biotechnol
Agilent 1260 HPLC system using a ProPac WCX-10 column, 4.0 ×250 mm (Dionex, CA, USA). Mobile phase A was 10 mM Na 2 HPO 4 , pH = 7.5, and mobile phase B was 100 mM NaCl in 10 mM Na2HPO4, pH=7.5. A linear gradient elution from 30 to 75 % B in 20 min was used to separate mAb charge variant with detection at 280 nm. The flow rate was 1 mL/min, and the separation was performed at room temperature. Acidic variants and basic variants were defined as represented by the peaks that were eluted earlier or later than the main peak. Reverse-transcript PCR (RT-PCR) and real-time PCR (qRT-PCR) analysis RT-PCR and qRT-PCR analyses were employed to detect and quantify the transcription level of basic Cps (CpB, CpH, CpM) in CHO cells. Total cell RNA was extracted using TRIzol Reagent (Invitrogen, CA, USA) and then reverse transcribed by MMLV Reverse Transcriptase (Promega, NY, USA). RT-PCR was performed using the cDNA on a 2720 Thermal Cycler (Applied Biosystems, Singapore) with conditions as follows: 5 min at 94 °C, 30 cycles of 30 s at 94 °C, 25 s at 58 °C (53 °C for CpH), and 30 s at 72 °C, followed by 10 min at 72 °C for a final extension. GAPDH, a housekeeping gene, was used as an internal control. The primers used in this study are summarized in Table 2. The amplification products were analyzed using 1.5 % agarose gel with ethidium bromide. qRT-PCR reaction was performed and monitored using the SYBR Green PCR Master mix and the ABI Prism7700 Sequence Detection System (Applied Biosystems, CA, USA). The relative levels of basic Cps mRNAs were analyzed using the 2−ΔΔCt method with GAPDH as an internal control. Intracellular pH (pHi) measurement pHi was determined as previously described (deZengotita et al. 2002). Briefly, cells were incubated with the dye carboxyseminaphthorhodafluor-1 (carboxy-SNARF-1) (Invitrogen, CA, USA) for 30 min. The ratio of the fluorescence emission at 585 and 640 nm was then measured using a FACS Calibur flow cytometer (Becton Dickinson, CA, USA). Additionally, standards were generated by incubating cells in
Table 2
K + buffers with different pH values while nigericin (Invitrogen, CA, USA) was added to induce the equilibration of the intracellular and extracellular pH. The pHi of sample was then calculated from the fluorescence ratio of standard curve. Intracellular amino acid extraction and concentration determination Intracellular amino acids were extracted as previously described (Dietmair et al. 2010) with slight modifications. Briefly, cell fluid (5×106 cells) was quickly mixed with cold quenching solution (0.9 % (w/v) NaCl, 0 °C) (v/v, 1:4). After centrifugation (1000 g for 1 min at 0 °C), cell pellets were separated and resuspended in 1-mL ice-cold 50 % aqueous acetonitrile. The mixture was vortexed and incubated on ice for 10 min. After centrifugation (20000g for 5 min at 0 °C), supernatant was separated and dried by a SpeedVac (Thermo Savant, NY, USA). The dried extracts were resuspended in water for further analysis. Concentrations of intracellular and extracellular amino acids were quantified using an Agilent 1260 HPLC coupled with a Nova-Pak C18 column, 3.9×150 mm (Waters, MA, USA), following the precolumn derivation using AccQ·Fluor reagent kit (Waters, MA, USA). Intracellular amino acid concentrations were calculated based on cell volume by presuming that each cell was spherical with determined diameter (Hansen and Emborg 1994). Cell-free assay The batch culture under LL condition (Table 1) was designated as the cell process (CP) and performed as described. On day 6, 100-mL cell fluid (CF) from CP was centrifuged (1000 rpm for 5 min) to discard cells and sterile filtered (0.22 μm) as harvested cell fluid (HCF). The HCF was then divided into four parts and added additional Arg and Lys, i.e., three parts with additional 8 mM Arg, 8 mM Lys, 8 mM Arg and Lys, respectively, and the remaining part with no addition. The four HCFs were then reincubated under the same condition of CP for another 8 days. Samples were taken daily from the HCFs to determine the lysine variant level during the incubation.
Primers used for analysis of basic Cps in CHO cells
Gene
Forward primers (5′–3′)
Reverse primers (5′–3′)
Product length (bp)
GAPDH CpB CpH CpM
AGGTTGTCTCCTGCGACTTCA CAGCGAGAATCCACAACTCA TGTGCTCTCCGCCAATCT CCGATACCACCACCAGGAAG
GAGGTCCACCACTCTGTTGCT ATCAATGTTGACCACAGGCA AACTACTGTCATCGTCGTTCTT CCGAGTGCTGTTGATTAGTTGT
143 265 203 305
Appl Microbiol Biotechnol
For the kinetic study of C-terminal lysine processing during the cell culture process, four batch cultures with various Arg and Lys concentrations (Table 1) were performed as described above. A 50-mL CF was taken on days 1, 3, 5, 6, 7, 8 from each condition, prepared as HCF and reincubated under the same condition of CP. Samples were taken daily from the HCFs to determine the lysine variant level during the incubation.
Results Cell growth, mAb production, and lysine variant level To determine the potential roles of Arg and lysin C-terminal lysine processing during the cell culture process, batch cultures with various Arg and Lys concentrations were performed as illustrated in Table 1. Cultures were performed in triplicate. Figure 1 shows the profiles of cell growth, mAb production, and lysine variant level. No statistically significant difference of cell growth and mAb production was found in cultures with various concentrations of Arg and Lys (Fig. 1a, b). The maximum viable cell densities reached 3.5– 4.2×106 cells/mL on day 4 under the four conditions (Fig. 1a). Meanwhile, the accumulated mAb concentrations reached 211–235 mg/L during the 8-day culture (Fig. 1b). Contrary to the similar cell growth and mAb production, significant differences of lysine variant levels were observed under the four conditions on day 8 (Fig. 1c). Specifically, only 18.7 % lysine variant level was observed when both Arg and Lys concentrations were low (2 mM Arg and 2 mM Lys, defined as LL). However, when 10 mM Arg or Lys was added (LH or HL), the lysine variant level increased significantly to 28.3 or 23.6 % (p=0.003, 0.015), and even reached up to 31.8 % in the case of both high Arg and Lys concentrations (HH). These variations indicate a positive effect of Arg and Lys on mAb lysine variant level. Moreover, it can also be speculated that Lys has obviously stronger influence than Arg (p=0.014). Since heterogeneity of C-terminal lysine residues is believed to result from varying degree of proteolysis by basic Cps, enzyme level, pH conditions, and product (Arg and Lys) inhibition, which potentially affect the enzymatic reaction, were then investigated under various Arg and Lys conditions to elucidate this positive effect of Arg and Lys on mAb lysine variant level. Basic Cps detection by RT-PCR assay In this study, the presence of normal basic Cps in CHO cells (including CpB, CpH, and CpM) was first examined by RTPCR assay. The sample was taken from the batch culture under LL condition (Table 1) on day 6, and the result is shown in Fig. 2a. The lengths of the PCR products of approximately 260 bp (CpB) and 200 bp (CpH) obtained from the
Fig. 1 Profiles of a cell growth, b mAb production, and c lysine variant level during the batch cultures with various Arg and Lys concentrations. Black square LL, black circle LH, black triangle HL, black diamond HH. Note that the error bars indicate the standard deviations from three independent experiments, and p values are estimated by two-tailed Student’s t test
electropherogram were in accordance with the designed values (Table 2), thus indicating the presence of CpB and CpH in the CHO cell line we used. However, no apparent signal for CpM was obtained. The transcription levels of CpB and CpH during the batch cultures with various Arg and Lys concentrations were further investigated by qRT-PCR, and the results are shown in Fig. 2b, c. As can be seen, the relative mRNA levels (to GAPDH) of CpB and CpH were stable throughout the 8-day culture, and no significant difference existed among the four conditions, indicating that the transcription levels of CpB and CpH should not account for the different lysine variant levels.
Appl Microbiol Biotechnol
Fig. 3 Profile of pHi during the batch cultures with various Arg and Lys concentrations. Black square LL, black circle LH, black triangle HL, black diamond HH. Note that the values presented are the average from two independent experiments
pHi (no more than 0.05 unit) was believed not to significantly affect the activity of basic Cps (Bradley et al. 1996; Wolff et al. 1962). Additionally, extracellular pH (pHe) was also measured during the batch cultures, and no significant difference was observed among the four conditions (data not shown), thus indicating that the different lysine variant levels should not be attributed to pH condition.
Amino acid concentration
Fig. 2 RT-PCR and qRT-PCR analysis of basic Cps in CHO cells with GAPDH as an internal control. a Typical agarose gel electropherogram for the detection of CpB, CpH, and CpM, with a 50-bp DNA ladder. Relative b CpB and c CpH mRNA levels during the batch cultures with various Arg and Lys concentrations. The mRNA levels for each gene were normalized to LL condition on day 1. □: LL, : LH, : HL, ■: HH. Note that the error bars indicate the standard deviations from three independent experiments
Variations in intracellular pH (pHi) The pHi values during the batch cultures with various Arg and Lys concentrations are monitored and shown in Fig. 3. It was found that the pHi values decreased gradually throughout the culture processes regardless of conditions. From an initial value of 7.46, the pHi decreased to 7.40 (LL), 7.36 (LH), 7.37 (HL), and 7.35 (HH) on day 8 under the four conditions. A lower pHi value was observed throughout the culture process under higher Arg and Lys concentrations (HH) when compared with the other conditions. However, this difference in
As the catalytic products of basic Cps, Arg and Lys were reported to have the potential to inhibit the activity of basic Cps in a dose-dependent manner (Hook 1990; Wolff et al. 1962). Therefore, intracellular and extracellular amino acid concentrations were quantified to determine whether these two basic amino acids reached the inhibition concentration and consequently affected the lysine variant level. Samples under the four culture conditions were taken every 2 days, and the results are presented in Fig. 4. For an initial concentration of 2 or 10 mM in the medium, extracellular Arg concentration decreased to approximately 1 or 9 mM on day 7 (Fig. 4a), whereas intracellular concentration increased to approximately 2 or 9 mM on day 3 and then decreased to approximately 1 or 4 mM on day 7 (Fig. 4c). Similarly, for an initial concentration of 2 or 10 mM in the medium, extracellular Lys concentration decreased to approximately 0.5 or 7 mM on day 7 (Fig. 4b), whereas intracellular concentration increased to approximately 1 or 4 mM on day 3 and then decreased to approximately 0.5 or 3 mM on day 7 (Fig. 4d). Altogether, the intracellular and extracellular Arg and Lys concentrations ranged from 0.5 to 2 mM for lower initial concentration (2 mM). For higher initial concentration (10 mM), the intracellular and extracellular Arg and Lys concentrations ranged from 3 to 10 mM, which were about 4-fold higher than those for lower concentration.
Appl Microbiol Biotechnol
Fig. 4 a Extracellular Arg, b extracellular Lys, c intracellular Arg, and d intracellular Lys concentrations during the batch cultures. □: LL, : LH, : HL, ■: HH. Note that the error bars indicate the standard deviations from three independent experiments
Determined by the published works, the Ki values (inhibition constant) of CpB from porcine were 0.5 and 13 mM for Arg and Lys, respectively (Wolff et al. 1962), and the Ki values of CpH from bovine were 4.6 and 7.6 mM for Arg and Lys, respectively (Hook 1990). In the present study, the intracellular and extracellular Arg and Lys concentrations ranged from 0.5 to 10 mM during the batch cultures and were the same order of magnitude (mM) as the reported Ki values, indicating the possibility of occurrence of product inhibition effect of Arg and Lys on basic Cps. Therefore, the significant differences in intracellular and extracellular concentrations of the two basic amino acids under the four conditions, as discussed above, would lead to different extents of inhibition, thus resulting in different lysine variant levels. However, the inhibition effect of Arg and Lys on basic Cps in CHO cells should be demonstrated with direct evidence. Moreover, the C-terminal lysine processing during the culture process should be further dynamically characterized. Cell-free assay In order to confirm the inhibition effect of Arg and Lys on basic Cps in CHO cells, a cell-free assay of HCF incubation was performed with various Arg and Lys concentrations.
Lysine variant levels during the incubation with various concentrations of Arg and Lys are determined and shown in Fig. 5a. It was found that the lysine variant level decreased from 15.6 to 8.2 % in the end, at an average rate of 0.9 %/day in the absence of additional Arg and Lys, indicating the presence of extracellular basic Cps in the HCF and its activity of C-terminal lysine cleavage. However, when additional 8 mM Lys, 8 mM Arg, or 8 mM Arg and Lys were added, the lysine variant levels decreased from 15.6 to 9.7 %, 9.4 or 10.4 % in the end, at a lower rate of 0.7 %/day, 0.8 %/day, or 0.6 %/day compared with the sample with no addition, demonstrating that the activity of extracellular basic Cps was inhibited by these two basic amino acids. Furthermore, a linear regression of the logarithm of lysine variant level (%) against incubation time shows that the extracellular C-terminal lysine cleavage process is a first-order reaction with the rate coefficient (k) represented by the fitted slope (Fig. 5b). Therefore, the extracellular process is significantly inhibited by Arg and Lys because the k value decreased by 37 % with additional 8 mM Arg and Lys when compared with the sample with no addition. Moreover, a greater reduction (25 %) in k value with additional 8 mM Lys than that (21 %) with equivalent Arg also indicates that Lys is a more effective inhibitor of basic Cps in CHO cells.
Appl Microbiol Biotechnol
various Arg and Lys concentrations are also determined and shown in Fig. 6c. As under the LL condition, the k values also increased under the other conditions (LH, HL, HH) during the culture process. Moreover, higher concentrations of Arg and Lys led to a lower k value, which should be attributed to the inhibition effect of the two basic amino acids.
Discussion
Fig. 5 Time course of HCF lysine variant level during incubation with various Arg and Lys concentrations. a Profile of lysine variant level. b Plot of the lysine variant level on a log scale. Black square no additional Arg and Lys, black circle 8 mM additional Lys, black triangle 8 mM additional Arg, black diamond 8 mM additional Arg and Lys. Note that the values presented are the average from three independent experiments
C-terminal lysine processing is a complicated process that occurs both intracellularly and extracellularly. To further characterize the C-terminal lysine processing, we conducted a kinetic study of the C-terminal lysine processing during the cell culture process. As shown in Fig. 6a, lysine variant level of CP under LL condition increased from 8.9 % on day 1 to 17.7 % on day 9. Contrarily, the lysine variant levels of HCFs gradually decreased correspondingly owing to the extracellular basic Cps activity. By plotting the logarithm of lysine variant level (%) against culture time (Fig. 6b), the k values of the extracellular C-terminal lysine processing during the batch culture are estimated and illustrated in Fig. 6c. An increase in the k value from 0.052 day−1 on day 1 to 0.086 day−1 on day 8 (condition LL) indicates that the rate of extracellular Cterminal lysine cleavage would increase during the culture process since it positively correlates with the k value (Fig. 6c) and lysine variant level (Fig. 6a). Therefore, considering that the measured lysine variant level in the CP increased during the culture process (Fig. 6a) regardless of the increased extracellular lysine cleavage rate (Fig. 6c), the lysine variant level of mAb newly secreted in the CP would also increase probably because of a reduction in the intracellular cleavage rate. Additionally, the k values of the batch cultures with
Arg and Lys, two basic amino acids, have been well characterized by their noticeable roles in cell growth and mAb production during the cell culture process (Carrillo-Cocom et al. 2014; Gonzalez-Leal et al. 2011). They were also found to stabilize mAbs during the purification and formulation process (Falconer et al. 2011). However, their effect on mAb quality, especially charge variants, has been rarely discovered. In this study, to obtain a detailed understanding of their potential roles in the C-terminal lysine processing during cell culture process, a typical antibody-producing CHO cell line was employed and cultivated with various Arg and Lys concentrations. Contrary to the unaffected cell growth and mAb production, of particular interest finding was that the lysine variant levels varied significantly with various Arg and Lys concentrations. A higher level of lysine variant was detected in the culture with higher concentrations of Arg and Lys, a phenomenon which is the first time to report to our best knowledge. As an enzymatic process, C-terminal lysine cleavage is potentially affected by various factors, such as temperature, pH, substrate and enzyme levels, and inhibition and activation effects. Accordingly, possible causes for the elevated lysine variant levels under higher Arg and Lys concentration conditions were further investigated. In this study, RT-PCR assay was performed to detect the basic Cps in CHO cells. Previous Western blot results and CHO genome database data indicated the possible presence of CpB, CpM, and CpH in CHO cells (Luo et al. 2012). Our RT-PCR results demonstrated the transcription of both CpB and CpH, confirming the presence of these two basic Cps in CHO cells. However, no apparent signal for CpM was found, which might be attributed to the low expression level in the CHO cell line we used. Furthermore, the transcription levels of CpB and CpH during the batch cultures were determined and showed no significant difference among the cultures with various concentrations of Arg and Lys (Fig. 2b, c). The transcription process is typically regarded as the dominant factor controlling protein expression (Jiang et al. 2006), and RT-PCR is widely used for relative and quantitative measurement of gene expression (Wong et al. 2010a; Wong et al. 2010b). The present result revealed that high concentrations of Arg and Lys in culture medium had no obvious influence on the transcription of CpB and CpH in CHO cells. However, the actual
Appl Microbiol Biotechnol
Fig. 6 Kinetic study of C-terminal lysine processing during the batch cultures with various Arg and Lys concentrations. a Time course of CP and HCF lysine variant level during incubation under LL condition. b Plot of the lysine variant level of HCF during incubation on a log scale. c k values of extracellular C-terminal lysine cleavage during the batch cultures with various Arg and Lys concentrations. a, b Black diamond
CP, black square HCF on day 1, black circle HCF on day 3, black triangle HCF on day 5, white square HCF on day 6, white circle HCF on day 7, white triangle HCF on day 8. c black square LL, black circle LH, black triangle HL, black diamond HH. Note that the values presented are the average from three independent experiments
activity of CpB and CpH might be affected by other factors that correlate more with Arg and Lys. The potential difference in pHi and pHe caused by various levels of Arg and Lys was another concern which might influence the C-terminal lysine processing by affecting the activity of basic Cps. Exist as zwitterions at physiological pH, amino acids act as buffers of pHi (Chen and Harcum 2005; Edwards et al. 1998). High levels of Arg and Lys, two basic amino acids with pI values of 10.76 and 9.74, respectively, had the potential to raise the pHi and therefore influence the activity of basic Cps. Surprisingly, a slight decrease (no more than 0.05 unit) in pHi was observed when cultivated with higher levels of Arg and Lys (Fig. 3). This decrease in pHi might be caused by the increased accumulation of acidic metabolite (e.g., lactic acid), which might outweigh the opposite effect caused by the higher levels of Arg and Lys (Jeong et al. 2006). Nevertheless, such a difference in pHi was believed not to significantly affect the activity of basic Cps (Bradley et al. 1996; Wolff et al. 1962).
On the other hand, the activity of basic Cps might be competitively inhibited by Arg and Lys depending on the inhibitor concentration. Actually, product inhibition effects of CpB and CpH are ubiquitous and play important roles in biological active peptide synthesis (Hook and LaGamma 1987; Wolff et al. 1962). Arg and Lys, the catalytic products of basic Cps, have been previously found to effectively inhibit the activity of basic Cps at the mM order of magnitude (Hook and LaGamma 1987; Wolff et al. 1962). In the present study, remarkable intracellular and extracellular Arg and Lys concentrations were observed comparable with Ki values of basic Cps. Although the Ki values were from bovine and porcine, they were of the same order as those from other species due to the high degree of homology of basic Cps from different species (Bradley et al. 1996; Hook 1988). Therefore, the present results suggested the possible product inhibition of basic Cps by Arg and Lys during the cell culture process. Accordingly, the different lysine variant levels (Fig. 1c) could be attributed to the different inhibition effects caused by various
Appl Microbiol Biotechnol
concentrations of Arg and Lys. Furthermore, in this study, Lys showed as a more effective inhibitor than Arg in inhibiting the C-terminal lysine cleavage (Fig. 1c). However, according to the reported study, the Ki values of the two Cps from multiple sources are generally smaller for Arg than those for Lys, indicating that Arg is a more effective inhibitor (Hook and LaGamma 1987; Wolff et al. 1962). These inconsistent results indicate the possible difference in the mechanism of product inhibition within different species, which is worth further investigation and verification. Moreover, a series of cell-free assays of HCF incubation were performed to confirm the product inhibition effect of Arg and Lys in CHO cells and to further characterize the Cterminal lysine processing during the culture process. Cellfree assay seems to be a powerful approach to simplify complicated PTMs, including C-terminal lysine processing, which occur both intracellularly and extracellularly (Dick et al. 2008; Luo et al. 2012). The significant inhibition effect of Arg and Lys on the extracellular basic Cps was verified using the kinetic analysis. As expected, the kinetic analysis also revealed that the extracellular C-terminal lysine processing is a firstorder reaction, which is consistent with the lysine processing of IgG2 in vivo (Cai et al. 2011). Moreover, the intracellular C-terminal lysine processing might also be a first-order reaction on the basis of the mechanism of the enzymatic reaction. Additionally, a kinetic study of lysine variant level during the cell culture process enabled thorough understanding of the Cterminal lysine processing. It was found that the extracellular cleavage rate increased while the intracellular cleavage rate decreased during the culture process, which might be attributed to the distinct intracellular and extracellular variations of various factors, such as enzyme, cofactor, and inhibitor levels, in addition to Arg and Lys concentrations. Although the inhibition effects of Arg and Lys on the intracellular basic Cps remain unconfirmed, the inhibition effect on the extracellular C-terminal lysine processing clearly plays an important role in modulating the final lysine variant level. Besides the mAb used in this study, we have also investigated the product inhibition effect on other IgGs, and it seems to be an un-neglectable issue during typical CHO cell cultures. More importantly, much attention should be paid to these two basic amino acids because of their influence on both productivity and quality of mAb during culture process. Maintaining the concentrations of Arg and Lys at a low level will be an effective approach for reducing the lysine variant level. However, simply reducing the concentration in medium may cause growth arrest and subsequent low mAb concentration, since Arg and Lys are two essential amino acids in cell culture process with demonstrated positive impact on cell growth and mAb production (Carrillo-Cocom et al. 2014; GonzalezLeal et al. 2011). Therefore, optimization of culture process, especially the medium component, under the guidance of quality by design (QbD) seems to be a better approach to
alleviate the product inhibition of Arg and Lys (Rouiller et al. 2012). Via the balance of various amino acids and other nutrients, the reduced lysine variant level can be achieved, and the consistency of the culture process can also be well controlled. In conclusion, high concentrations of Arg and Lys significantly enhanced mAb lysine variant level by product inhibition effect on basic Cps in CHO cell cultures. This observation and elucidation of the impact of specific amino acids on lysine variant level help to better understand and control product heterogeneity in mAb manufacturing.
Acknowledgments This work was supported by the National Natural Science Foundation of China (No. 21406066, 21206040), the National High Technology Research and Development Program of China (863 Program) (No. 2012AA02A303). Conflict of interest The authors have no conflict of interest.
References Aghamohseni H, Ohadi K, Spearman M, Krahn N, Moo-Young M, Scharer JM, Butler M, Budman HM (2014) Effects of nutrient levels and average culture pH on the glycosylation pattern of camelidhumanized monoclonal antibody. J Biotechnol 186:98–109 Bradley G, Naudé RJ, Muramoto K, Yamauchi F, Oelofsen W (1996) Ostrich (Struthio camelus) carboxypeptidase B: Purification, kinetic properties and characterization of the pancreatic enzyme. Int J Biochem Cell Biol 28(5):521–529 Cai B, Pan H, Flynn GC (2011) C-terminal lysine processing of human immunoglobulin G2 heavy chain in vivo. Biotechnol Bioeng 108(2):404–412 Carrillo-Cocom LM, Genel-Rey T, Araiz-Hernandez D, Lopez-Pacheco F, Lopez-Meza J, Rocha-Pizana MR, Ramirez-Medrano A, Alvarez MM (2014) Amino acid consumption in naive and recombinant CHO cell cultures: producers of a monoclonal antibody. Cytotechnology. doi:10.1007/s10616-014-9720-5 Chen P, Harcum SW (2005) Effects of amino acid additions on ammonium stressed CHO cells. J Biotechnol 117(3):277–286 Crowell CK, Grampp GE, Rogers GN, Miller J, Scheinman RI (2007) Amino acid and manganese supplementation modulates the glycosylation state of erythropoietin in a CHO culture system. Biotechnol Bioeng 96(3):538–549 deZengotita VM, Schmelzer AE, Miller WM (2002) Characterization of hybridoma cell responses to elevated pCO2 and osmolality: intracellular pH, cell size, apoptosis, and metabolism. Biotechnol Bioeng 77(4):369–380 Dick LW Jr, Qiu D, Mahon D, Adamo M, Cheng KC (2008) C-terminal lysine variants in fully human monoclonal antibodies: investigation of test methods and possible causes. Biotechnol Bioeng 100(6): 1132–1143 Dietmair S, Timmins NE, Gray PP, Nielsen LK, Krömer JO (2010) Towards quantitative metabolomics of mammalian cells: Development of a metabolite extraction protocol. Anal Biochem 404(2):155–164 Dorai H, Ganguly S (2014) Mammalian cell-produced therapeutic proteins: heterogeneity derived from protein degradation. Curr Opin Biotechnol 30:198–204
Appl Microbiol Biotechnol Edwards LJ, Williams DA, Gardner DK (1998) Intracellular pH of the mouse preimplantation embryo: amino acids act as buffers of intracellular pH. Hum Reprod 13(12):3441–3448 Falconer RJ, Chan C, Hughes K, Munro TP (2011) Stabilization of a monoclonal antibody during purification and formulation by addition of basic amino acid excipients. J Chem Technol Biotechnol 86(7):942–948 Fan Y, Jimenez Del Val I, Muller C, Wagtberg Sen J, Rasmussen SK, Kontoravdi C, Weilguny D, Andersen MR (2015) Amino acid and glucose metabolism in fed-batch CHO cell culture affects antibody production and glycosylation. Biotechnol Bioeng 112(3):521–535 Gonzalez-Leal IJ, Carrillo-Cocom LM, Ramirez-Medrano A, LopezPacheco F, Bulnes-Abundis D, Webb-Vargas Y, Alvarez MM (2011) Use of a Plackett-Burman statistical design to determine the effect of selected amino acids on monoclonal antibody production in CHO cells. Biotechnol Prog 27(6):1709–1717 Greene D, Das B, Fricker LD (1992) Regulation of carboxypeptidase E. Effect of pH, temperature and Co2+ on kinetic parameters of substrate hydrolysis. Biochem J 285:613–618 Hansen H, Emborg C (1994) Extra-and intracellular amino acid concentrations in continuous Chinese hamster ovary cell culture. Appl Microbiol Biotechnol 41(5):560–564 Harris RJ (1995) Processing of C-terminal lysine and arginine residues of proteins isolated from mammalian cell culture. J Chromatogr A 705(1):129–134 Hook V, LaGamma E (1987) Product inhibition of carboxypeptidase H. J Biol Chem 262(26):12583–12588 Hook VY (1988) Regulation of carboxypeptidase H by inhibitory and stimulatory mechanisms during neuropeptide precursor processing. Cell Mol Neurobiol 8(1):49–55 Hook VY (1990) Arginine and lysine product inhibition of bovine adrenomedullary carboxypeptidase H, a prohormone processing enzyme. Life Sci 47(13):1135–1139 Jeong DW, Cho IT, Kim TS, Bae GW, Kim IH, Kim IY (2006) Effects of lactate dehydrogenase suppression and glycerol-3-phosphate dehydrogenase overexpression on cellular metabolism. Mol Cell Biochem 284(1-2):1–8 Jiang Z, Huang Y, Sharfstein ST (2006) Regulation of recombinant monoclonal antibody production in chinese hamster ovary cells: a comparative study of gene copy number, mRNA level, and protein expression. Biotechnol Prog 22(1):313–318 Jing Y, Borys M, Nayak S, Egan S, Qian Y, Pan S-H, Li ZJ (2012) Identification of cell culture conditions to control protein aggregation of IgG fusion proteins expressed in Chinese hamster ovary cells. Process Biochem 47(1):69–75
Liu H, Gaza-Bulseco G, Faldu D, Chumsae C, Sun J (2008) Heterogeneity of monoclonal antibodies. J Pharm Sci 97(7):2426– 2447 Luo J, Zhang J, Ren D, Tsai WL, Li F, Amanullah A, Hudson T (2012) Probing of C-terminal lysine variation in a recombinant monoclonal antibody production using Chinese hamster ovary cells with chemically defined media. Biotechnol Bioeng 109(9):2306–2315 Ponniah G, Zhang H-M, Nowak C, Neill A, Gonzalez-Lopez N, Patel R, Chang G, Kita A, Andrien B, Liu H (2014) In vitro and in vivo modifications of recombinant and human IgG antibodies. mAbs 6(5):1–11 Rouiller Y, Solacroup T, Deparis V, Barbafieri M, Gleixner R, Broly H, Eon-Duval A (2012) Application of Quality by Design to the characterization of the cell culture process of an Fc-Fusion protein. Eur J Pharm Biopharm 81(2):426–437 Schenerman MA, Sunday BR, Kozlowski S, Webber K, GazzanoSantoro H, Mire-Sluis A (2004) CMC strategy forum report. BioProc Int 2(2):42–52 Sun Y-t, Zhao L, Ye Z, Fan L, Liu X-p, Tan W-S (2013) Development of a fed-batch cultivation for antibody-producing cells based on combined feeding strategy of glucose and galactose. Biochem Eng J 81:126–135 Vlasak J, Ionescu R (2008) Heterogeneity of monoclonal antibodies revealed by charge-sensitive methods. Curr Pharm Biotechnol 9(6): 468–481 Wolff EC, Schirmer E, Folk J (1962) The kinetics of carboxypeptidase B activity I. Kinetic parameters. J Biol Chem 237(10):3094–3099 Wong DC, Wong NS, Goh JS, May LM, Yap MG (2010a) Profiling of Nglycosylation gene expression in CHO cell fed-batch cultures. Biotechnol Bioeng 107(3):516–528 Wong NS, Wati L, Nissom PM, Feng HT, Lee MM, Yap MG (2010b) An investigation of intracellular glycosylation activities in CHO cells: effects of nucleotide sugar precursor feeding. Biotechnol Bioeng 107(2):321–336 Yang M, Butler M (2000) Enhanced erythropoietin heterogeneity in a CHO culture is caused by proteolytic degradation and can be eliminated by a high glutamine level. Cytotechnology 34(1-2):83–99 Yu XC, Borisov OV, Alvarez M, Michels DA, Wang YJ, Ling V (2009) Identification of codon-specific serine to asparagine mistranslation in recombinant monoclonal antibodies by high-resolution mass spectrometry. Anal Chem 81(22):9282–9290 Yuk IH, Russell S, Tang Y, Hsu WT, Mauger JB, Aulakh RP, Luo J, Gawlitzek M, Joly JC (2015) Effects of copper on CHO cells: Cellular requirements and product quality considerations. Biotechnol Prog 31(1):226–238
