Serum Dependence (Molecular Biology)

A large number of serum-free media for tissue culture of cell lines have been developed (1, 2), and many of these are now commercially available, but the bulk of cell culture on the laboratory scale is still carried out in the presence of serum. Serum provides the cells with growth factors, attachment factors, hormones, nutrients, minerals, and trypsin inhibitor activity. It also buffers the medium against pH fluctuations and traces of toxins. On the other hand, it is a poorly defined natural product that carries the risk of contamination with viruses and mycoplasma, is variable in quality and constitution, is sometimes in limited supply, and adds complexity to the isolation of cell products from the medium. Over the past 20 years, several recipes have been generated for the replacement of serum, usually with defined supplements containing selenium, insulin, transferrin, steroids, and lipids, and often with additional minerals and intermediary metabolites. The main advantages of these serum-free recipes are definition, reproducibility, and selectivity for the cell type that may be propagated. Serum dependence is also a feature that is most strongly expressed in normal cells, as one of the recognized features of Neoplastic Transformation is a reduction in the requirement for serum, reflecting the increase in autonomous growth control typical of transformed cells.

1. Constitution of Serum

Data on serum analysis are traditionally derived from clinical laboratories and depend on conventional analyses and assays based on the major constituents, and those that show fluctuations that are of interest in diagnosis of disease or nutritional deficiency. It is now possible, however, to obtain data from serum suppliers, with greater emphasis on those constituents of greater significance to cell culture, and whose determination usually requires immunoassays or functional bioassays.


The major constituents of serum are listed in Table 1. There are undoubtedly other constituents that may be of importance, but these are the main ones that have been identified. Those that have been shown to be required in serum-free medium, in addition to the regular constituents found in serum-containing medium, are given in boldface type. The major roles of these constituents are listed in Table 2.

Table 1. Constituents2 of Serum

Concentration Range of Each b

Albumin

20-50 mg/mL

Amino acids

tmp214-33

Calcium

tmp214-34

Chlorides

tmp214-35

Cholesterol

tmp214-36

Fatty acids

tmp214-37

Fetuin

In FB only

Fibronectin

tmp214-38

Globulins

1—15 mg/mL

Glucose

0.6—1.2 mg/ml

Growth Factors: EGF, PDGF, IGF,FGF,

1—100 ng/mL

IL-1, IL-6, Insulin

Hydrocortisone

10—200 nM

Hexosamine

0.0—1.2 mg/mL

Highest in

human

Iron

tmp214-39

Lactic acid

0.5—2.0 mg/mL

Highest in

FB

Linoleic acid

tmp214-40

Phospholipids

0.7—3.0 mg/mL

Polyamines putrescine,spermidine

tmp214-41

Potassium

5—15 mM

Proteinase inhibitors:a1-antitrypsin, a2-

0.5—2.5 mg/mL

macroglobulin

Protein, total

40—80 mg/mL

Pyruvic acid

2—10 |g/mL

Selenium

tmp214-42

Sodium

135—155 mM

Total lipids

2—10 mg/mL

Transferrin

2—4 mg/mL

Tri-iodotyrosine

Urea

tmp214-43

Vitamins

tmp214-44

Vitamin A

10—100 ng/mL

Zinc

tmp214-45

a Constituents inbold type are those that have been used to supplement serum-freemedia. b Concentrationrange is very approximate and only intended to convey the order ofmagnitude.

Table 2. Major Functions of Serum in Cell Culture

Function

Constituent

Mitogenic activity

Growth factors, platelet-derived growth factor(PDGF) and insulin-like growth factors (IGF-1,IGF-2)

Attachment factors

Fibronectin, fetuin

Antagonize trypsin used in subculture

a1-Antitrypsin,a2-macroglobulin

Bind toxins

Albumin

Detoxify free radicals

Selenium

Stimulate nutrient uptake, eg,

Insulin

glucose and amino acids

Bind and transport iron

Transferrin

Provide intermediary metabolites

Pyruvate, a-ketoglutarate, adenosine, etc

Viscosity to buffer cells against

Albumin

mechanical damageduring manipulation

Many of the growth factors in serum are derived from platelets; hence serum from naturally clotted blood is usually superior to serum separated from the cellular constituents by centrifugation. The growth factors released by platelets include platelet-derived growth factor (PDGF) and transforming growth factor b (TGFb), which have effects that are cell-type-specific. PDGF is a mitogen for mesodermally derived cells, such as fibroblasts, and glial cells, such as astrocytes; TGFb is cytostatic for many epithelial cells. Hence, the presence of serum is likely to favor fibroblastic cells, rather than epithelium, in cultures from normal tissues, and this has added weight to the argument in favor of using serum-free selective media to grow normal epithelial cells. Transformed cells may be exceptional, as they often produce a spectrum of autocrine growth factors, giving them a degree of autonomous control over mitogenesis. In addition, they may also produce TGFa, which reverses the cytostatic effect of TGFb and gives it a mitogenic and transformation-like effect. The inhibitory effect of TGFb can also be reduced by the presence of a feeder layer of 3T3 primitive embryonic mouse mesodermal cells, which appears to degrade or block activation of TGFb in cocultures with epithelium. Such feeder layers are frequently used to generate cultures of epithelial cells, such as epidermal keratinocytes (3), in serum-containing medium.

Growth of cells in the absence of serum frequently requires modification of the substrate to enable cell attachment and, for normal cells at least, the cell spreading that is necessary for cell proliferation. This modification can be a simple chemical treatment with, for example, poly-D-lysine, which neutralizes the negatively charged plastic and produces a slight positive charge. The incorporation of spermidine or putrescine into the medium (4) may play a similar role. Other protocols require extracellular matrix constituents, such as collagen, fibronectin, or laminin (5), which will interact with specific integrin receptors on the cell surface and promote adhesion to the substrate. In many cases, the cells may be capable of generating these matrix constituents themselves, as is suggested by the improved survival of cells reseeded into the flask from which they were released by trypsin treatment, or on to a substrate conditioned by other cells, such as vascular endothelium (6). It has been suggested that serum contains many such attachment factors that condition the substrate, modify its charge, and allow attachment of extracellular matrix molecules important in cell adhesion. Fibronectin is known to be present in serum, although in a modified form, and may participate in this.

2. Effect of Transformation

Reduced serum dependence is frequently used as a criterion for the identification of transformed cells. While normal cells will block at a restriction point in the G1 phase of the cell cycle in the absence of serum and will re-enter cycle when serum is restored (7, 8), transformed cells will tend to progress into S phase regardless of the serum concentration, although they may arrest at later points in the cycle because of nutritional deficiencies. This distinction is created by the reduced dependence of transformed cells on exogenous growth factors (see Neoplastic Transformation) as a result of the expression of autonomous control mechanisms, such as autocrine growth factor production, for example, PDGF b-subunit in gliomas (9), or modifications in signal transduction leading to unregulated, permanently active induction of mitogenesis, such as ras mutations in colon carcinoma (10) (see Oncogenes, Oncoproteins).

Transformed cells are also able to proliferate without the degree of cell spreading required with normal cells, even to the extent of proliferating in suspension. Hence, there is reduced dependence on attachment factors present in serum for cells to initiate proliferation. Transformed cells are not devoid of cell adhesion molecules (CAMs), but they frequently have modifications affecting the selectivity of attachment and, possibly more important, alterations to the extracellular domain, which modify the interactions with the actin cytoskeleton, which, in turn, may make redundant cell spreading and the formation of adhesion plaques.

3. Serum-Free Media

Although serum supplementation is still widely used, there are several problems associated with the use of serum, most of which derive from its variability and undefined nature. Although a fairly precise analysis can be performed of the major constituents of serum, it is the minor constituents, those present in small amounts, but with high specific activity, that are difficult to detect and most likely to be variable. Some of these are listed in Table 1 and include hormones and growth factors. Although batch testing minimizes this problem, each batch has a limited shelf life, and it is unlikely that the next batch will be identical. Hence, although the nutritional content and physical properties of serum can be reproduced fairly accurately, the signaling content cannot, giving rise to considerable physiological variation from batch to batch. With the dramatic increase in the use of cell culture by the pharmaceutical industry, acute worldwide shortages have been avoided only by the development of low serum or serum-free formulations by drug companies, many of which, unfortunately, are not available in the public domain. In order to meet the requirements of Good Laboratory Practice (The United Kingdom Compliance Programme: Department of Health, London, 1989) and Good Manufacturing Practice (Medicines Control Agency, "Rules and Guidance for Pharmaceutical Manufacturers and Distributors 1997," The Stationery Office Ltd., London, 1977), pharmaceutical companies have also been obliged to address the problems of viral contamination, which most research laboratories still choose to ignore. As the elimination of viruses from serum has proved to be extremely difficult, the obvious alternative is serum-free media, which have the additional advantages of minimizing the risk of mycoplasma infections and reducing the contamination of cell products with serum proteins, thereby facilitating downstream processing.

One of the major advantages of serum-free media has been the development of media that are selective for individual cell types. The range is now quite extensive, and many are available commercially from companies such as Bio Whittaker, Sigma, Invitrogen (Gibco), and ICN. Some examples are given in Table 3; for a more extensive list see Barnes et al. (11), Freshney, (12), Davis, (13) and Jayme and Gruber (2).

Table 3. Serum-FreeSelective Media

Medium

Cell Type

Reference

MCDB 105

Human fibroblasts

McKeehan, et al. (1977) InVitro 13, 399; Bettger et al.(1981) Proc. Natl. Acad. Sci. USA 78 5588

MCDB 110

WI38, MRC5,

IMR90

MCDB 202

Chicken fibroblasts

McKeehan et al. (1977), InVitro 13, 399

MCDB 402

Mouse 3T3 cells

Shipley and Ham, (1981) InVitro 17, 656

MCDB 411

Mouse

Agy, et al. (1981) In Vitro 17, 671

neuroblastoma

C1300

MCDB 153

Human

Boyce and Ham, (1983) J.

keratinocytes

Invest.Dermatol. 81, 33s

MCDB 170

Mammary

Hammond et al. (1984) Proc. Natl.Acad.

epithelium

Sci. USA 81, 5435

Iscove’s

Hemopoietic cells

Iscove and Melchers (1978) J. Exp.Med.

147, 923

LHC

Bronchial

Lechner and LaVeck (1985) J. Tissue Cult.

epithelium

Meth. 9, 43

HITES

Small cell lung

Carney et al. (1981) Proc. Natl.Acad. Sci.

cancer

USA 78, 3185

WAJC 404

Prostatic epithelium Chaproniere and McKeehan (1986) Cancer

Res. 46, 819

DMEM:F12,

Various, with

Sato (1979) in Methods inEnzymology, W.

1:1

supplementation

B. Jakoby and I. H. Pastan, eds.,

AcademicPress, New York, pp. 94-109

Unfortunately, a move to serum-free medium is not without its difficulties. Individual recipes may be required for each cell type maintained, and the time required to develop media for specific cell types is quite significant, possibly several years. The problem is diminishing with an increase in the supply of serum-free media from commercial sources, but it is still substantially more expensive than serum-containing media and may be beyond the reach of laboratories on modest budgets. There are also problems with subculture, as serum is normally responsible for the inhibition of trypsin that is used to liberate cells when they are reseeded. Trypsin damage can be reduced by using purified trypsin at a reduced temperature (14) and/or by incorporating a trypsin inhibitor in the medium at reseeding.

The major additions to medium to replace serum are insulin, selenium, and iron-saturated transferrin, and these are to be found in most serum-free formulations. In addition, hydrocortisone, cholera toxin, or isoprenaline are often added to increase proliferation of epithelial cells. Hydrocortisone probably acts by increasing attachment via the induction of proteoglycans, which may also activate cytokines and growth factors (15, 16), cholera toxin and isoprenaline increase the intracellular concentration of cyclic AMP, which is mitogenic in some epithelial cells. Lipid precursors, such as cholesterol and linoleic acid, ethanolamine, and phosphoethanolamine, high density lipoproteins (HDL), and crude lipid preparations, such as soya bean lipid, are often included and may contribute to the biosynthesis of membranes or second messengers. Thiol compounds such as b-mercaptoethanol are often used to inhibit oxidative stress induction from free radicals. In addition to insulin and hydrocortisone, other growth factors and hormones include epidermal growth factor (EGF), fibroblast growth factor (FGF), insulin-like growth factor-1 (IGF-1, somatomedin C) and -2 (IGF-2, multiplication stimulating activity (MSA)) (17), PDGF, follicle-stimulating hormone (FSH), prolactin, tri-iodotyrosine, estradiol, and prostaglandins such as PGF2a;. In addition to selenium and iron, other trace elements include copper, zinc, manganese, molybdenum, tin, nickel, vanadium, and silicon. Finally, in the absence of the detoxifying effect of serum proteins, there is a requirement for highly purified reagents and water.

Next post:

Previous post: