The Problem with Serum

Transport of lipids, proteins, drugs, hormones, and many other molecules in vivo is done through serum, and its transport protein serum albumin. Hydrophobic nutrients have traditionally been provided to in vitro cultures by supplementation with animal serum, or the carrier protein serum albumin.  Lipids and fatty acids must be “carried” to cells and delivered in small amounts, since they are hydrophobic in free form, and produce “soap” micelles at certain concentrations, which are toxic to cells1.  Serum from cows, horses, goats, and humans are currently used for delivery of fats and lipids, as well as numerous other nutrients.  This is a black box approach at best, as the serum from a cow or other donor contains signaling molecules, nutrients, drugs, and hormones that are appropriate for the donor; not necessarily appropriate for an individual cell type from another species.

Serum albumin has the capacity to carry at least seven different compounds, including an average of two fatty acids.  They are complex in their binding behaviors and extremely variable in their natural cargo contents2.  In vivo, serum, and serum albumin are active carriers of proteins and cytokines as well as lipids. These are potent factors in cellular health and behavior in vitro as well. 

Additionally, the differences between the needs of a specific cell type being cultured and the collective needs of an entire organism are enormous.   Serum and serum albumin contain nutrients of general importance like lipids and trace elements, as well as numerous proteins and growth factors that may not be appropriate for a given individual cell type2.  In vivo, the contents of the serum are carried through many systems and cells of all types accept the nutrients that are appropriate for them, producing metabolites that may be taken up by serum albumin.  When serum is used in cell culture media, a wide range of proteins and factors are all available for interaction with cells, often causing unintended variability in experimental results by releasing content that inappropriately signals cells, or potentially lacking content specific for the cultured cell type.  Because serum and serum albumin content vary so greatly, amounts and types of cytokines in fetal bovine serum (FBS) and human serum (HS) are not reliably published.  Contents are variable depending on the health of the source organism, and this variability is carried over into experimental results.   Inter-species variation is also enormous. HS commonly contains about four times more cholesterol than FBS and significant differences occur in other lipid levels as well5.    Published averages illustrate the large differences in cytokine content between human and bovine serum (Table 1), and the variability of bovine serum content.

Human SerumBovine Serum
Insulin like Growth Factor 176 ng/mL111 ng/mL
Transforming Growth Factor B147 ng/mL12.6 ng/mL
Platelet Derived Growth Factor5 ng/mL1-100 ng/mL
Epithelial Growth Factor1.5 ng/mL1-100 ng/mL
Hepatocyte Growth Factor0.8 ng/mLundetected
Vascular Endothelial Growth Factor0.063 ng/mL1-100 ng/mL
Fibroblast Growth Factor0.0019 ng/mL37.3 ng/mL

Table 1. Average growth factor contents in human and bovine serum. Amounts are extremely variable. Human serum data3 Bovine serum data4

Variability has long been recognized as a problem, but recently the greater issue of contamination has come to the forefront.   Serum and serum albumin may also contain viruses, prions and other bioactive materials that can cause contamination of cultures, and in turn, the eventual consumers of cell based products and stem cells6.

In today’s advanced cell applications, cells and cell products are being applied directly to humans, or potentially being consumed as food.   In order to assure the safety of patients and consumers, absence of contamination must first be established.   Unfortunately, when a cell line has been exposed to animal derived products, complete absence of deadly contaminants like prions cannot be assured and testing for the plethora of other possible contaminants possibly present in serum and serum albumin is costly.  This is a major roadblock to some of the most promising therapies and cellular applications actively in development.  

When used in cell culture, serum albumin delivers its cargo readily to cell membranes.  When a biological moiety is delivered, the loss allosterically effects the binding affinities of the other binding sites on albumin and leaves an open site for uptake of other media components7.   This makes the variability of sera contents only one aspect of its unpredictability as a lipid carrier, with contaminants and uptake of needed nutrients adding to the potential problems.

As cell research, and especially induced pluripotent stem cell (iPSC) research has progressed, there has been a move to “chemically-defined” media conditions6.   This means culture conditions which do not include animal or human sourced proteins or tissue extracts, including serum and albumin. Currently, the most widely used chemically-defined media is Essential 88.  This media has been very successful in supporting stem cells in a pluripotent state, but is very low in lipid content, containing only 150nM linoleic acid in its base, Dulbecco’s Modified Eagle Medium (DMEM).  Lack of lipids in chemically-defined media has been found to induce de novo lipid synthesis, which modifies metabolism, reducing mitochondrial respiration and altering cell behavior9,10. Cells can synthesize a wide range of lipids, but capabilities and requirements vary considerably between cell types.  Therefore it seems optimal to provide the specific lipids and fatty acids required for the needs of the specific cells being cultured, while avoiding others, as it has been found that fatty acids are not only structural molecules and energy sources for cells, but also can be potent signaling molecules11,12 and post-translational modifiers of vital proteins13,14.

References Cited

  1. Alsabeeh, N., B. Chausse, P. A. Kakimoto, A. J. Kowaltowski and O. Shirihai Cell culture models of fatty acid overload: Problems and solutions Biochimica Et Biophysica Acta-Molecular and Cell Biology of Lipids 1863(2): 143-151 (2018)
  2. Fasano, M., S. Curry, E. Terreno, M. Galliano, G. Fanali, P. Narciso, S. Notari and P. Ascenzi The extraordinary ligand binding properties of human serum albumin. Iubmb Life 57(12): 787-796  (2005)
  3. Rauch, C., E. Feifel, E. M. Amann, H. P. Spotl, H. Schennach, W. Pfaller and G. Gstraunthaler Alternatives to the Use of Fetal Bovine Serum: Human Platelet Lysates as a Serum Substitute in Cell Culture Media Altex-Alternatives to Animal Experimentation 28(4): 305-316(2011)
  4. Fresney, I. R. Culture of Animal Cells: A Manual of Basic Technique and Specialized Applications 112-120 (John Wiley & Sons, Inc. 2011)
  5. Haylett, A. K. and J. V. Moore Comparative analysis of foetal calf and human low density lipoprotein: relevance for pharmacodynamics of photosensitizers. Journal of Photochemistry and Photobiology B-Biology 66(3): 171-178 (2002)
  6. Gishto, N. D. P. R. A. Human embryonic stem cell cultivation: historical perspective and evolution of xeno free culture systems. Reproductive biology and endocrinology 13(1): 9 (2015)
  7. Fasano, M., S. Curry, E. Terreno, M. Galliano, G. Fanali, P. Narciso, S. Notari and P. Ascenzi The extraordinary ligand binding properties of human serum albumin. Iubmb Life 57(12): 787-796 (2005)
  8. Beers, J., D. R. Gulbranson, N. George, L. I. Siniscalchi, J. Jones, J. A. Thomson and G. Chen Passaging and colony expansion of human pluripotent stem cells by enzyme-free dissociation in chemically defined culture conditions. Nature Protocol7(11): 2029-2040 (2012)
  9. Lamaziere, A., D. Farabos, C. Wolf and P. J. Quinn The deficit of lipid in cultured cells contrasted with clinical lipidomics. Molecular Nutrition & Food Research 57(8): 1401-1409 (2013)
  10. Lin, Z. Y., F. Liu, P. L. Shi, A. Y. Song, Z. Huang, D. Y. Zou, Q. Chen, J. X. Li and X. Gao Fatty acid oxidation promotes reprogramming by enhancing oxidative phosphorylation and inhibiting protein kinase C. Stem Cell Research & Therapy 9(47) (2018)
  11. Sumida, C. and E. A. Nunez “Fatty-Acids and cell signal transduction.” Journal of Lipid Mediators and Cell Signalling 9(2): 91-116 (1994)
  12. Papackova, Z. and M. Cahova Fatty Acid Signaling: The New Function of Intracellular Lipases. International Journal of Molecular Sciences 16(2): 3831-3855 (2015)
  13. Nile, A. H. and R. N. Hannoush Fatty acylation of Wnt proteins. Nature Chemical Biology 12(2): 60-69 (2016)
  14. McDonnell, E., S. B. Crown, D. B. Fox, B. Kitir, O. R. Ilkayeva, C. A. Olsen, P. A. Grimsrud and M.D. Hirschey Lipids Reprogram Metabolism to Become a Major Carbon Source for Histone Acetylation. Cell Reports 17(6): 1463-1472(2016)