A 2015 study by Kishishita and co-workers optimized a chemically defined feed for monoclonal antibody (mAb) production in Chinese hamster ovary (CHO) cells by supplementing the amino acids most depleted during culture. Supplementing serine, cysteine, and tyrosine raised the day-14 antibody titer by about 60% and coincided with a switch from lactate production to lactate consumption, and a soluble Cys-Tyr-Ser tripeptide reproduced the benefit without the osmolality increase caused by the acidic free-amino-acid feed. The mechanisms behind these observations were left open: why the supplementation promotes the lactate-consumption switch; how amino-acid effects on the specific glucose consumption rate can be separated from the confounding effect of elevated osmolality; how the tripeptide is utilized intracellularly (it escaped detection by amino-acid analysis); under what conditions and concentration thresholds threonine, proline, and glycine protect CHO cells from ammonium stress; and why 25 mM and 50 mM tripeptide elicit different productivity and lactate responses. All five questions remain open. Later work clarified the general mechanism of the lactate shift (lactate dehydrogenase isoenzyme expression and NAD+/NADH redox), showed that dipeptides are imported intact and hydrolyzed intracellularly, and found other nutrient routes to lactate consumption (TCA-cycle intermediates, NAD+ precursors), but no study answers any of the five questions for the serine-cysteine-tyrosine system, and at least one industrial study found serine supplementation increasing lactate when glucose is abundant.

Background

Chinese hamster ovary (CHO) cells are the predominant mammalian host for commercial production of therapeutic monoclonal antibodies (mAbs), most commonly grown in fed-batch culture, in which a concentrated nutrient feed is added incrementally to an initially complete batch culture. Modern processes use chemically defined media, meaning formulations in which every component has a known chemical identity and concentration; this avoids the regulatory risk and lot-to-lot variability of serum and hydrolysates. Two waste metabolites constrain such processes. Lactate, produced from glucose-derived pyruvate by lactate dehydrogenase (LDH) when glycolytic flux exceeds mitochondrial oxidative capacity, accumulates to growth-inhibiting levels, and the base added to neutralize the resulting acidity raises the osmolality (the total dissolved-solute concentration), which itself suppresses cell growth while often increasing cell-specific productivity. Ammonium, generated mainly by catabolism of the amine groups of glutamine and asparagine, inhibits growth and alters protein glycosylation. A desirable and much-studied phenomenon is the lactate shift: many cultures switch in the stationary phase (after growth ceases) from net lactate production to net lactate consumption; Mulukutla, Gramer and Hu (2012) attributed this shift to growth-phase-dependent changes in LDH isoenzyme expression (the lactate-forming LDH-A decreasing relative to LDH-B) coupled to NAD+/NADH redox balance, where NAD+ is the oxidized cofactor that glycolysis must regenerate to keep running. Independently of mechanism, Chen and Harcum (2005) had reported that adding certain amino acids can partially mitigate the growth inhibition of ammonium-stressed CHO cells. Feed design is typically driven by spent-media analysis: amino acids that become depleted during a control culture are added to the feed at levels calculated from the depletion profiles. A practical bottleneck is tyrosine, whose poor solubility at neutral pH forces feed preparation at about pH 2, and the base needed to neutralize this acidic feed upon addition raises culture osmolality by roughly 200 mOsm/kg; short peptides (dipeptides, and later tripeptides — chains of two or three amino acids) dissolve far better and had been used to deliver tyrosine and other amino acids. In this setting, Kishishita et al. (2015) supplemented a three-fold concentrated chemically defined feed for a DHFR-deficient DXB11 CHO line (a subline lacking dihydrofolate reductase, used for gene amplification) expressing a single IgG. Adding the five most-depleted amino acids including asparagine (35.0 mM) and glutamine (84.0 mM) accumulated about 30 mM ammonium and lowered titer; adding only serine (50.0 mM), cysteine (1.8 mM), and tyrosine (14.5 mM) raised the day-14 titer about 60% above the unsupplemented control and the cultures switched to lactate consumption (lactate below about 2 g/L versus above 3 g/L in control). A chemically synthesized cysteine-tyrosine-serine (Cys-Tyr-Ser) tripeptide dissolved at up to 50 mM without any pH reduction, and 25 mM of it raised titer by about 35% with no osmolality increase, while 50 mM behaved differently. The paper quantified the specific glucose consumption rate (QGluc, glucose consumed per cell per unit time) but could not attribute its changes to amino acids versus osmolality. Regarding the mechanism of the lactate-consumption switch, the source paper speculated about an effect on glycolytic flux, in analogy to reported effects of AKT1 (a serine/threonine-protein kinase in growth-factor signaling that also stimulates glucose uptake and glycolysis) activity on glucose metabolism, but did not investigate it. Five mechanistic questions arising from this study and its cited literature form the present problem: the mechanism of the lactate-consumption switch; the osmolality confound on QGluc; the intracellular utilization pathway of the tripeptide (as reported by the source, undetectable by amino-acid analysis, an instrument that quantifies only free amino acids, with the authors calling for metabolic flux analysis, the isotope-tracer-based quantification of intracellular reaction rates); the conditions and concentration thresholds under which threonine, proline, and glycine protect CHO cells from ammonium, following the protection reported by Chen and Harcum (2005), cited by the source; and the concentration-dependent differential response between 25 and 50 mM tripeptide.

Problem Statement

Establish the mechanistic basis - metabolic, transport-related, and concentration-dependent - for the effects of serine, cysteine, and tyrosine (Ser/Cys/Tyr) supplementation, delivered either as free amino acids or as the chemically synthesized Cys-Tyr-Ser tripeptide, in chemically defined CHO fed-batch mAb production, as observed by Kishishita et al. (2015). Five mechanistic questions arise from that study: (1) By what mechanism does Ser/Cys/Tyr supplementation promote the switch from lactate production to lactate consumption during the stationary phase (supplemented cultures kept lactate below about 2 g/L after the switch, versus above 3 g/L in the unsupplemented control)? (2) How much of the change in specific glucose consumption rate (QGlucQ_{Gluc}) is attributable to the amino acids themselves and how much to the elevated osmolality that accompanies their delivery - a separation the source study could not achieve experimentally? (3) How is the Cys-Tyr-Ser tripeptide processed intracellularly: is it hydrolyzed extracellularly or taken up intact, and what is the metabolic fate of its residues? (The tripeptide was not detectable by amino-acid analysis.) (4) Under what conditions - within the ammonium range of the source system (roughly 10-50 mM, which includes the ~30 mM accumulated ammonium) and its chemically defined medium - and above what concentration thresholds can threonine, proline, and glycine protect CHO cells from ammonium toxicity (that is, measurably mitigate the ammonium-induced inhibition of growth or titer), or is there no such protective threshold in that regime? The protection concept originates with Chen and Harcum (2005), cited by the source study; the source study reports that maintaining these amino acids above 20 mM did not mitigate the toxicity of about 30 mM accumulated ammonium in its experiments. (5) Why do 25 mM and 50 mM Cys-Tyr-Ser tripeptide supplementation elicit different productivity and lactate-production responses, as reported by the source study, despite similar viable cell density, glucose, QGlucQ_{Gluc}, and ammonium profiles? Each of these five questions is separately open and none is resolved by the others; all five must be resolved to establish the mechanistic basis sought above.

References

  1. Shohei Kishishita, Satoshi Katayama, Kunihiko Kodaira, Yoshinori Takagi, Hiroki Matsuda, Hiroshi Okamoto, Shinya Takuma, Chikashi Hirashima, Hideki Aoyagi (2015). Optimization of chemically defined feed media for monoclonal antibody production in Chinese hamster ovary cells. Journal of Bioscience and Bioengineering 120(1):78-84. DOI: 10.1016/j.jbiosc.2014.11.022. https://doi.org/10.1016/j.jbiosc.2014.11.022 (source paper)
  2. Peifeng Chen, Sarah W. Harcum (2005). Effects of amino acid additions on ammonium stressed CHO cells. Journal of Biotechnology 117(3):277-286. DOI: 10.1016/j.jbiotec.2005.02.003. https://doi.org/10.1016/j.jbiotec.2005.02.003
  3. Bhanu Chandra Mulukutla, Michael Gramer, Wei-Shou Hu (2012). On metabolic shift to lactate consumption in fed-batch culture of mammalian cells. Metabolic Engineering 14(2):138-149. DOI: 10.1016/j.ymben.2011.12.006. https://doi.org/10.1016/j.ymben.2011.12.006
  4. Jun Luo, Natarajan Vijayasankaran, Jennifer Autsen, Rodell Santuray, Terry Hudson, Ashraf Amanullah, Feng Li (2012). Comparative metabolite analysis to understand lactate metabolism shift in Chinese hamster ovary cell culture process. Biotechnology and Bioengineering 109(1):146-156. DOI: 10.1002/bit.23291. https://doi.org/10.1002/bit.23291
  5. Jong Kwang Hong, Shilpa Nargund, Meiyappan Lakshmanan, Sarantos Kyriakopoulos, Do Yun Kim, Kok Siong Ang, Dawn Leong, Yuansheng Yang, Dong-Yup Lee (2018). Comparative phenotypic analysis of CHO clones and culture media for lactate shift. Journal of Biotechnology 283:97-104. DOI: 10.1016/j.jbiotec.2018.07.042. https://doi.org/10.1016/j.jbiotec.2018.07.042
  6. Xiaolin Zhang, Rubin Jiang, Henry Lin, Sen Xu (2020). Feeding tricarboxylic acid cycle intermediates improves lactate consumption and antibody production in Chinese hamster ovary cell cultures. Biotechnology Progress 36(4):e2975. DOI: 10.1002/btpr.2975. https://doi.org/10.1002/btpr.2975
  7. Jernej Horvat, Mojca Narat, Oliver Spadiut (2020). The effect of amino acid supplementation in an industrial Chinese Hamster Ovary process. Biotechnology Progress 36(5):e3001. DOI: 10.1002/btpr.3001. https://doi.org/10.1002/btpr.3001
  8. Cécile Toussaint, Olivier Henry, Yves Durocher (2016). Metabolic engineering of CHO cells to alter lactate metabolism during fed-batch cultures. Journal of Biotechnology 217:122-131. DOI: 10.1016/j.jbiotec.2015.11.010. https://doi.org/10.1016/j.jbiotec.2015.11.010
  9. James Morrissey, Ayca Cankorur-Cetinkaya, Luigi Grassi, Annie J. Harwood-Stamper, Jonathan Welsh, Cleo Kontoravdi (2025). Nicotinamide reverses the Warburg effect in Chinese hamster ovary cell culture. bioRxiv preprint. DOI: 10.1101/2025.06.18.660349. https://doi.org/10.1101/2025.06.18.660349
  10. Andres Sánchez-Kopper, Max Becker, Jennifer Pfizenmaier, Christian Kessler, Andreas Karau, Ralf Takors (2016). Tracking dipeptides at work-uptake and intracellular fate in CHO culture. AMB Express 6:48. DOI: 10.1186/s13568-016-0221-0. https://doi.org/10.1186/s13568-016-0221-0
  11. Natascha Verhagen, Andy Wiranata Wijaya, Attila Teleki, Muhammad Fadhlullah, Andreas Unsöld, Martin Schilling, Christoph Heinrich, Ralf Takors (2020). Comparison of l-tyrosine containing dipeptides reveals maximum ATP availability for l-prolyl-l-tyrosine in CHO cells. Engineering in Life Sciences 20(9-10):384-394. DOI: 10.1002/elsc.202000017. https://doi.org/10.1002/elsc.202000017
  12. Seo-Young Park, Jinsung Song, Dong-Hyuk Choi, Uiseon Park, Hyeran Cho, Bee Hak Hong, Yaron R. Silberberg, Dong-Yup Lee (2024). Exploring metabolic effects of dipeptide feed media on CHO cell cultures by in silico model-guided flux analysis. Applied Microbiology and Biotechnology 108(1):123. DOI: 10.1007/s00253-023-12997-0. https://doi.org/10.1007/s00253-023-12997-0
  13. Brian J. Kirsch, Sandra V. Bennun, Adam Mendez, Amy S. Johnson, Hongxia Wang, Haibo Qiu, Ning Li, Shawn M. Lawrence, Hanne Bak, Michael J. Betenbaugh (2022). Metabolic analysis of the asparagine and glutamine dynamics in an industrial Chinese hamster ovary fed-batch process. Biotechnology and Bioengineering 119(3):807-819. DOI: 10.1002/bit.27993. https://doi.org/10.1002/bit.27993
  14. Iman Shahidi Pour Savizi, Nader Maghsoudi, Ehsan Motamedian, Nathan E. Lewis, Seyed Abbas Shojaosadati (2022). Valine feeding reduces ammonia production through rearrangement of metabolic fluxes in central carbon metabolism of CHO cells. Applied Microbiology and Biotechnology 106(3):1113-1126. DOI: 10.1007/s00253-021-11755-4. https://doi.org/10.1007/s00253-021-11755-4
  15. Fatemeh Torkashvand, Behrouz Vaziri, Shayan Maleknia, Amir Heydari, Manouchehr Vossoughi, Fatemeh Davami, Fereidoun Mahboudi (2015). Designed amino acid feed in improvement of production and quality targets of a therapeutic monoclonal antibody. PLoS ONE 10(10):e0140597. DOI: 10.1371/journal.pone.0140597. https://doi.org/10.1371/journal.pone.0140597
  16. Bhagya S. Yatipanthalawa, Yih Yean Lee, Sally L. Gras, Gregory J.O. Martin (2026). Amino acid metabolism, demand and supply in Chinese Hamster ovary cell culture - A comprehensive literature review. Biotechnology Advances 89:108866. DOI: 10.1016/j.biotechadv.2026.108866. https://doi.org/10.1016/j.biotechadv.2026.108866
  17. Hossein Kavoni, Iman Shahidi Pour Savizi, Nathan E. Lewis, Seyed Abbas Shojaosadati (2025). Recent advances in culture medium design for enhanced production of monoclonal antibodies in CHO cells: a comparative study of machine learning and systems biology approaches. Biotechnology Advances 78:108480. DOI: 10.1016/j.biotechadv.2024.108480. https://doi.org/10.1016/j.biotechadv.2024.108480
  18. Yang Wu, Yuxin Liu, Yingnan Si, Alexander Noonan, Yang Wang, Vincent Morrison, Christa Beaureagard, Valerie Besset, Bing Wang, Jianlin Xu, Henry Lin (2026). Cysteine-condensed thiazolidine as an effective cysteine source for CHO cell culture. Journal of Biotechnology 419:125-135. DOI: 10.1016/j.jbiotec.2026.08.007. https://doi.org/10.1016/j.jbiotec.2026.08.007