Fatty acid and oxylipin concentration differ markedly between different fetal bovine serums: A cautionary note

Abbreviations ARA arachidonic acid DHA docosahexaenoic acid FAME fatty acid methyl ester FBS fetal bovine serum GC-FID gas chromatography flame ionization detection HEPE hydroxy eicosapentaenoic acid HETE hydroxy eicosatetraenoic acid HUFA highly unsaturated fatty acid LA linoleic acid LC-MS liquid chromatography mass spectrometry LOX lipoxygenase PUFA polyunsaturated fatty acid SPM specialized pro-resolving lipid mediator INTRODUCTION

Fetal bovine serum (FBS) has been used in cell culture for more than half a century. Initially introduced by Theodore Puck at the end of the 1950s (Puck et al., 1958), it turned out that FBS can be used as a universal supplement in cell culture as it contains essential substances for cell homeostasis and proliferation (van der Valk et al., 2018). However, the use of FBS shows poor reproducibility and comparability between laboratories (Baker, 2016; Honn et al., 1975; van der Valk & Gstraunthaler, 2017). Although alternatives such as human platelet lysate or chemically defined serum-free medium are available since 1980s (Burnouf et al., 2016; Gstraunthaler, 2003; Witzeneder et al., 2013), they are not commonly used, often only suitable for single-cell lines, and are believed to be less efficient than FBS regarding, for example, cell proliferation or viability (Bauman et al., 2018; Gstraunthaler, 2003; Piletz et al., 2018). Hence, FBS is still the “gold standard” in cell culture (Piletz et al., 2018). Plenty of studies were carried out regarding the high variability of FBS component concentrations in different batches (Baker, 2016; Bauman et al., 2018; Esber et al., 1973; Honn et al., 1975; Zheng et al., 2006); however, only little is known about the serum lipid composition and its oxidation products (eicosanoids and other oxylipins). Besides influencing cell proliferation and morphology, the cells also utilize major and minor FBS components such as polyunsaturated fatty acids (PUFAs) and oxylipins. In order to ensure reliable and reproducible outcomes of cell culture studies regarding lipid biology, researchers should make sure that the fatty acid pattern and oxidation level are comparable within and between studies. In the present study, we determined the concentrations of free and total (free + esterified) oxylipins and their precursor fatty acids in six different FBS purchased from three companies to evaluate the PUFA and oxylipin variability and baseline levels in FBS. Our results demonstrate that even the use of two different batches from the same company could lead to different outcomes regarding the formation and biology of oxylipins.

MATERIAL AND METHODS

FBS products were purchased from Biowest (Nuaille, France), Merck (Darmstadt, Germany), and Biochrom (Berlin, Germany). In the following they are indicated as company 1 (1A, standardized; 1B non-standardized), company 2 (2A; 2B; non-standardized), and company 3 (3A; 3B; different standardized lots). Sample preparation as well as oxylipin and fatty acid determination was carried out as described (Kutzner et al., 2019; Ostermann et al., 2014; Ostermann et al., 2020; Rund et al., 2018). In brief, fatty acids were transesterified to fatty acid methyl esters (FAME) following lipid extraction with methyl tert-butyl ether/methanol. FAMEs were quantified by means of gas chromatography flame ionization detection (GC-FID) (Ostermann et al., 2014). Free oxylipins were extracted from FBS by protein precipitation with methanol. After solid phase extraction (SPE), free oxylipins were analyzed by means of liquid chromatography–mass spectrometry (LC-MS) (Kutzner et al., 2019; Rund et al., 2018). For the determination of total oxylipins, an additional hydrolysis step to liberate esterified oxylipins was included prior to SPE (Ostermann et al., 2020).

RESULTS AND DISCUSSION

All FBS products were analyzed for the fatty acid pattern by means of GC-FID (Table 1). The concentration of the highly unsaturated fatty acids (HUFA) as well as linoleic acid (LA) and α-linolenic acid showed pronounced differences between FBS of different companies (Table 1). For several fatty acids and batches such as company 1 and 2, the concentration of eicosapentaenoic acid (EPA), arachidonic acid (ARA), n3-ARA, dihomo-γ-linolenic acid, and LA are in the same range (e.g., for EPA 16 ± 2 μM/17 ± 1 μM vs. 17 ± 1 μM/15 ± 2 μM), while for Company 3, lower levels for all investigated fatty acids were determined in both batches. Interestingly, in the two “standardized” FBS batches purchased from company 3 the concentration for EPA, n3-docosapentaenoic acid, and docosahexaenoic acid (DHA) were not comparable as they were two-fold lower in batch 3B (e.g., for EPA 5 ± 1 μM) compared with 3A (11 ± 1 μM). It should be noted that the %n3 (48%–55%) and %n6 in HUFA (45%–52%) (Lands, 2008) varied only slightly between the FBS batches. The differences in fatty acid concentration can probably be explained by different animal fed as described for other cow-derived products such as milk and meat (Machado Neto et al., 2015; Mansbridge & Blake, 1997). Particularly the amount of protein-rich soy in the feed containing mostly LA could be a major factor influencing FBS fatty acid content. It seems that the fatty acid pattern in the FBS is not part of the standardization process of the manufactures. Taking into account that typically 10% of serum is added to the cell culture medium, this results in baseline concentrations in media of 13 μM n3-PUFA and 16 μM n6-PUFA for FBS 2A only due to the use of FBS. Because PUFAs play a key role in physiological homeostasis (Calder, 2012), the availability of fatty acids thus not only modulate fatty acid pattern (Gunn et al., 2017) but also influence the biological endpoints monitored in the cell culture experiments.

TABLE 1. Concentration of selected fatty acids in different commercial fetal bovine serum (FBS) Serum concentration (μM) 1A standardized 1B non-standardized 2A non-standardized 2B non-standardized 3A standardized 3B standardized 18:2n6 46 ± 4 49 ± 2 49 ± 3 50 ± 2 35 ± 3 29 ± 5 18:3n3 <1.8 <1.8 8.0 ± 0.8 8 ± 2 <1.8 <1.8 20:3n6 24 ± 1 24 ± 1 29 ± 1 26 ± 1 17.4 ± 0.2 15 ± 1 20:4n3 3.9 ± 0.2 5 ± 2 3.9 ± 0.3 5 ± 1 <1.6 <1.6 20:4n6 66 ± 3 65 ± 2 79 ± 2 70 ± 1 52 ± 3 40 ± 4 20:5n3 16 ± 2 17 ± 1 17 ± 1 15 ± 2 11 ± 1 5 ± 1 22:5n3 40 ± 1 34 ± 2 42 ± 1 29.5 ± 0.4 33 ± 2 19 ± 2 22:6n3 48 ± 2 49 ± 1 62 ± 3 45.78 ± 0.03 40 ± 2 26 ± 2 ∑ n3 PUFA 108 ± 2 104 ± 4 133 ± 4 102 ± 5 83 ± 6 50 ± 3 ∑ n6 PUFA 136 ± 8 138 ± 1 157 ± 4 147 ± 4 105 ± 6 85 ± 9 %n3 in HUFAa 55 ± 1 54 ± 1 54 ± 1 49 ± 1 54 ± 1 48 ± 2 %n6 in HUFAb 45 ± 1 46 ± 1 46 ± 1 51 ± 1 46 ± 1 52 ± 2 Note: Shown are concentrations (mean ± standard deviation, n = 3) of polyunsaturated fatty acids (PUFAs) in FBS from three different companies. Different batches of standardized (“superior/premium”) as well as non-standardized FBS were tested. Concentrations of fatty acids in serum were determined by means of gas chromatography flame ionization detection after lipid extraction and transesterification to fatty acids methyl esters. The lower limit of quantification was 1.4–2 μM. %n3 and %n6 in highly unsaturated fatty acids (HUFA: ≥20 carbon atoms, ≥3 double bonds) were calculated as described in Lands (2008). a %n3 in HUFA: 100*(20:4n3 + 20:5n3 + 22:5n3 + 22:6n3)/(20:3n6 + 20:4n6 + 20:4n3 + 20:5n3 + 22:5n3 + 22:6n3). b %n6 in HUFA: 100*(20:3n6 + 20:4n6)/(20:3n6 + 20:4n6 + 20:4n3 + 20:5n3 + 22:5n3 + 22:6n3).

One pathway how PUFAs elicit their biological effects is that they act as precursors of highly potent lipid mediators—eicosanoids and other oxylipins—regulating several (patho)physiological processes such as inflammation, vasoconstriction, or pain (Gabbs et al., 2015). The oxylipin concentrations differ considerably between the tested FBS (Figure 1). As expected, the highest concentrations were found for the 12-lipoxygenase (LOX) products 12-hydroxy eicosatetraenoic acid (HETE), 12-hydroxy eicosapentaenoic acid (HEPE), and 14-hydroxy docosahexaenoic acid (HDHA). During coagulation, platelets and subsequent 12-LOX are activated, leading to massively increased concentrations of these oxylipins in serum (Rund et al., 2020). For 12-LOX products, the difference in the concentrations between the serums is most pronounced, which might be explained by inconsistent serum generation procedures of the different companies.

image

Concentration of (a) free and (b) total oxylipins in fetal bovine serum (FBS) purchased from different companies (1–3). Shown are concentrations (mean ± standard deviation, n = 3) of a representative set of (a) free and (b) total (i.e., sum of free and esterified) oxylipins in FBS. Different batches of standardized (“superior/premium,” 1A, 3A, 3B) as well as non-standardized FBS (1B, 2A, 2B) were tested. Concentrations of oxylipins in serum were determined by means of LC–MS after protein precipitation, base hydrolysis in case of total oxylipins and solid phase extraction. Concentrations < lower limit of quantification are indicated by a dashed line. DiHDPE, dihydroxy docosapentaenoic acid; DiHETE, dihydroxy eicosatetraenoic acid; DiHETrE, dihydroxy eicosatrienoic acid; EpDPE, epoxy docosapentaenoic acid; EpETE, epoxy eicosatetraenoic acid; EpETrE, epoxy eicosatrienoic acid; HEPE, hydroxy eicosapentaenoic acid; HETE, hydroxy eicosatetraenoic acid; LOX, lipoxygenase; Tx, thromboxane; sEH, soluble epoxide hydrolase

A comparison of the hydroxy-PUFA of ARA and DHA shows that the HDHA concentrations of FBS 1A are similar compared with the HETE concentrations (e.g., 4-HDHA total 10.8 ± 0.5 nM as well as 5-HETE total 12.8 ± 0.4 nM) or even higher (e.g., 17-HDHA total 20.8 ± 0.6 nM and 15-HETE total 11.1 ± 0.6 nM), while the concentration of DHA (48 ± 2 μM) is lower compared with ARA (66 ± 3 μM). Thus, differences in the ratio of n3-PUFA and n6-PUFA derived oxylipins do not correlate with the ratio of the PUFA found in the serum. This may be explained by the fact that mammalian ALOX15 orthologs preferentially convert DHA to the corresponding hydro(pero)xy derivative (Kutzner et al., 2017). Comparing the two “standardized” FBS 3A and 3B reveals massive concentration differences for the total oxylipins, for example, for 5(R,S)-F2t-IsoP (0.57 ± 0.05 nM vs. 2.36 ± 0.02 nM), which is a well-known marker for fatty acid autoxidation (Rund et al., 2018). Additionally, 18-HEPE—which can also be formed by auto-oxidation—showed a four-fold higher concentration in the FBS 3B compared with 3A (1.88 ± 0.09 nM vs. 8.0 ± 0.4 nM). This could have distinct effect on the biology, since 18-HEPE is a precursor of the so-called specialized pro-resolving mediators (Serhan, 2014; Tjonahen et al., 2006).

Our results show that the commonly used 10% FBS in cell culture medium already leads to a considerable baseline level of PUFAs and oxylipins. This has to be taken into account when analyzing the effect of PUFAs or eicosanoids and other oxylipins in cell culture. Moreover, it is crucial to carefully analyze the used FBS regarding its fatty acid as well as oxylipin pattern and to select the most appropriate FBS supplier to ensure reliable and reproducible cell culture experiments.

ACKNOWLEDGMENTS

This study was supported by a PhD fellowship of the Ev. Studienwerk Villigst e.V. to Elisabeth Koch and by a grant of the German Research Foundation (DFG) to NHS (SCHE 1801).

ETHICS APPROVAL

In this work only commercially available biological materials were used.

CONFLICT OF INTEREST

The authors declare that they have no conflict of interest.

AUTHORS' CONTRIBUTIONS

EK and NHS designed experiments; EK, CH, LFF carried out experiments and measurements; EK and NHS wrote the manuscript; all authors approved the final manuscript.

REFERENCES

Baker M. Reproducibility: respect your cells! Nature. 2016; 537: 433– 5. Bauman E, Granja PL, Barrias CC. Fetal bovine serum-free culture of endothelial progenitor cells progress and challenges. J Tissue Eng Regen Med. 2018; 12: 1567– 78. Burnouf T, Strunk D, Koh MB, Schallmoser K. Human platelet lysate: replacing fetal bovine serum as a gold standard for human cell propagation? Biomaterials. 2016; 76: 371– 87. Calder PC. Mechanisms of action of (n-3) fatty acids. J Nutr. 2012; 142: 592S– 9S. Esber HJ, Payne IJ, Bogden AE. Variability of hormone concentrations and ratios in commercial sera used for tissue-culture. J Natl Cancer Inst. 1973; 50: 559– 62. Gabbs M, Leng S, Devassy JG, Monirujjaman M, Aukema HM. Advances in our understanding of oxylipins derived from dietary PUFAs. Adv Nutr. 2015; 6: 513– 40. Gstraunthaler G. Alternatives to the use of fetal bovine serum: serum-free cell culture. Altex-Altern Tierexp. 2003; 20: 275– 81. Gunn PJ, Green CJ, Pramfalk C, Hodson L. In vitro cellular models of human hepatic fatty acid metabolism: differences between Huh7 and HepG2 cell lines in human and fetal bovine culturing serum. Physiol Rep. 2017; 5:13532. Honn KV, Singley JA, Chavin W. Fetal bovine serum - multivariate standard. Proc Soc Exp Biol Med. 1975; 149: 344– 7. Kutzner L, Goloshchapova K, Heydeck D, Stehling S, Kuhn H, Schebb NH. Mammalian ALOX15 orthologs exhibit pronounced dual positional specificity with docosahexaenoic acid. BBA-Mol Cell Biol L. 2017; 1862: 666– 75. Kutzner L, Rund KM, Ostermann AI, Hartung NM, Galano JM, Balas L, et al. Development of an optimized LC-MS method for the detection of specialized pro-resolving mediators in biological samples. Front Pharmacol. 2019; 10: 169. Lands B. A critique of paradoxes in current advice on dietary lipids. Prog Lipid Res. 2008; 47: 77– 106. Machado Neto OR, Chizzotti ML, Ramos EM, Oliveira DM, Lanna DP, Ribeiro JS, et al. Fatty acid profile and meat quality of young bulls fed ground soybean or ground cottonseed and vitamin E. Animal. 2015; 9: 362– 72. Mansbridge RJ, Blake JS. Nutritional factors affecting the fatty acid composition of bovine milk. Br J Nutr. 1997; 78: 37– 47. Ostermann AI, Muller M, Willenberg I, Schebb NH. Determining the fatty acid composition in plasma and tissues as fatty acid methyl esters using gas chromatography - a comparison of different derivatization and extraction procedures. Prostaglandins Leukot Essent Fatty Acids. 2014; 91: 235– 41. Ostermann AI, Koch E, Rund KM, Kutzner L, Mainka M, Schebb NH. Targeting esterified oxylipins by LC-MS - effect of sample preparation on oxylipin pattern. Prostaglandins Other Lipid Mediat. 2020; 146:106384. Piletz JE, Drivon J, Eisenga J, Buck W, Yen S, McLin M, et al. Human cells grown with or without substitutes for fetal bovine serum. Cell Med. 2018; 10: 1– 11. Puck TT, Cieciura SJ, Robinson A. Genetics of somatic mammalian cells. 3. Long-term cultivation of EUPLOID cells from human and animal subjects. J Exp Med. 1958; 108: 945– 56. Rund KM, Ostermann AI, Kutzner L, Galano JM, Oger C, Vigor C, et al. Development of an LC-ESI(−)-MS/MS method for the simultaneous quantification of 35 isoprostanes and isofurans derived from the major n3- and n6-PUFAs. Anal Chim Acta. 2018; 1037: 63– 74. Rund KM, Nolte F, Doricic J, Greite R, Schott S, Lichtinghagen R, et al. Clinical blood sampling for oxylipin analysis - effect of storage and pneumatic tube transport of blood on free and total oxylipin profile in human plasma and serum. Analyst. 2020; 145: 2378– 88. Serhan CN. Pro-resolving lipid mediators are leads for resolution physiology. Nature. 2014; 510: 92– 101. Tjonahen E, Oh SF, Siegelman J, Elangovan S, Percarpio KB, Hong S, et al. Resolvin E2: identification and anti-inflammatory actions: pivotal role of human 5-lipoxygenase in resolvin E series biosynthesis. Chem Biol. 2006; 13: 1193– 202. van der Valk J, Gstraunthaler G. Fetal bovine serum (FBS) - a pain in the dish? Altern Lab Anim. 2017; 45: 329– 32. van der Valk J, Bieback K, Buta C, Cochrane B, Dirks WG, Fu J, et al. Fetal bovine serum (FBS): past - present - future. ALTEX. 2018; 35: 99– 118. Witzeneder K, Lindenmair A, Gabriel C, Holler K, Theiss D, Redl H, et al. Human-derived alternatives to fetal bovine serum in cell culture. Transfus Med Hemother. 2013; 40: 417– 23. Zheng XY, Baker H, Hancock WS, Fawaz F, McCaman M, Pungor E. Proteomic analysis for the assessment of different lots of fetal bovine serum as a raw material for cell culture. Part IV. Application of proteomics to the manufacture of biological drugs. Biotechnol Prog. 2006; 22: 1294– 300.

留言 (0)

沒有登入
gif