Tissue culture media and reagents*

STANDARDIZATION OF DIAGNOSTIC MATERIALS 5. Tissue culture media and reagents * F. T. PERKINSI The various factors that affect the growth of cell cultures in vitro are considered and suggestions are made as to where the standardization of components would help to provide more uniform products. The manufacture ofpowdered media in bulk and the establishment of cell banks have been major advances, but much research is neededfor the standardization or replacement of serum. Only ten years ago it would have been almost impossible to suggest what could be done towards the standardization of tissue culture media and reagents. However, much progress has been reported in recent years. The days of producing cell cultures from various tissues of animals, mainly monkeys taken from the jungle and subjected to a perfunctory quarantine period, are almost over. Many attempts have been made to standardize several factors inherent in the production of cell cultures and it is time that the findings of such studies were applied routinely. The part that cell cultures play in the diagnosis of diseases consists either in the isolation of a virus from a clinical specimen or in the determination of a parti- cular titre of antibody against a virus. In both these techniques the quality of the cell cultures used as the substrate is of importance. It is essential also that cell cultures used for virus vaccine production should be standardized in order to ensure consistency of pro- duction. The sensitivity of cell cultures to viruses may be affected by 3 main factors: the source of cells; the media; and the culture vessels. SOURCE OF CELLS Monkey kidney cells are an important adjunct to the diagnosis of disease. For the past 10 years the National Institute for Biological Standards and * Presented at the International Conference on Standar- dization, Atlanta, USA, 5-8 June 1973. 1 Head, Division of Immunological Products Control, National Institute of Biological Standards and Control, Hampstead, London NW3 6RB, England. Control, London, has been supplying primary rhesus monkey kidney cells to about 80 diagnostic laborato- ries, and human diploid cells to the public health laboratory service in the United Kingdom and to WHO virus laboratories throughout the world (Jacobs & Perkins, 1969). Although alternative tis- sues (such as human diploid cells of fetal origin) have been investigated, monkey cells retain their place in the diagnosis of viral infections. Manufacturers sup- plying cell monolayers or cell suspensions carry out controls in order to ensure that the cells are sensitive to viruses, but there is no uniform control of this activity. It is not suggested that each batch should be submitted to central control; however, it would be sensible to ensure that cells from different sources but originating from the same tissue had equal sensitivity to certain selected viruses. It is known that some cells propagated through a large number of cell doublings lose sensitivity to certain viruses. In this respect it is important to know the absolute number of cell doublings of the cell cultures supplied. Furthermore, it was not appre- ciated until recently how important it was to use cell cultures free from mycoplasmas. Such contaminants may have a marked effect on cell proliferation and enzyme activity (Rose et al., 1972); they may reduce the sensitivity of the cells to viruses; and many of them cause changes in the cell chromosomes (Paton et al., 1965). Here again, the use of a certified cell culture subjected to, and shown to comply with, specified tests is important. The ultimate in the standardization of cell cultures is the adoption of the principle of a cell bank. The practice of holding a large number of ampoules of 3075 -747 8 748 STANDARDIZATION cells, derived from the same cell source, in liquid nitrogen until such time as the cell bank has been shown to be suitable for the purposes for which it was established is being investigated on a wide scale. Human diploid cells (WI-38 and MRC5) (Hayflick & Moorhead, 1961, Jacobs et al., 1970) derived from fetal lung have been used for many years; diploid cell populations derived from other tissue (notably rabbit and monkey) (Wallace et al., 1972) are under investi- gation. One of the greatest problems with cell cultures is to ensure that they are not contaminated with myco- plasmas (McGarrity & Corniell, 1971) or viruses. Much work has been done on simian viruses and it is now known that any of 50 viruses may be present in a simian tissue (Hull, 1968). With a cell bank system, however, a portion of the bank can be propagated and exhaustively tested before use. It is known that passenger viruses in cell cultures can markedly affect the sensitivity of the tissue to some viruses. Such inherent contaminants must be eliminated. The establishment of a bank of cells has enabled much progress to be made in raising cell culture techniques from the realms of ritual to the level of systematic scientific studies. Thus the provision of a standard cell population, known to be free from contamination, has enabled the variables that may be present in media, sera, materials used for vessels, etc., to be investigated. A great advance has been the technique of mea- suring the antigenic profile of cells grown in vitro (Colombani et al., 1967; Schulman et al., 1964; Lapeyre et al., 1971).1 This technique allows the identification of cells and greatly assists in the detec- tion of an inadvertent admixture of cells, even of the same species. In addition, studies on the relative susceptibility of cells to a number of viruses may be a means of helping towards identifying cells and studies on enzyme patterns of cells may provide useful markers. CULTURE MEDIA In general the components of most cell culture media may be considered under 4 headings: (a) basic salts; (b) amino acids; (c) more complex chemicals (e.g., nucleic acids); and (d) serum. Chemical analysis is a great adjunct to the stan- dardization of components of known chemical com- position. There is little batch-to-batch variation in 1 See also van der Weerdt, C. M., Platelet antigens and isoimmunization, Amsterdam, 1965 (thesis). the chemicals manufactured by a well-tried process, and the only problem is not one of deficiency of a particular substance but toxicity of any by-products that may be present in the final product. This laboratory, for example, has made mixtures of amino acids that have given an analysis quantitatively similar to those of mixtures of the same amino acids, but from a different source; the mixture of amino acids from one source only has persistently failed to support the growth of cells. This may result either from the toxicity of a contaminant chemical or from the deficiency of any single amino acid. It may be explained also by the absence of an essential com- ponent (such as a heavy metal) other than an amino acid. An advance that has led to greater standardiza- tion of media is the production of large quantities of powdered medium thoroughly mixed and pretested for its suitability for the growth of cells before it is marketed (Hayflick et al., 1964; Jacobs, 1966). It is known that such a system has worked throughout the world in the propagation of human diploid fibroblast cells (WI-38) from fetal lung. These cells are rather fastidious in their growth requirements, and yet such a medium has been found to be suitable in all laboratories. The incorporation of organic buffers (HEPES) in order to maintain a constant pH for many days of growth of cells has made a further contribution towards standardization (Ceccarini & Eagle, 1971). These buffers prevent the frequent fluctuations in pH so commonly found in tissue culture work, and there is no doubt that the cells maintain a more uniform response to viruses because of this. For some time there has been considerable activity in the production of sterile powdered me- dium with or without sterile serum that is sterilized in the final container. In this way dissolving the powder in sterile water will provide medium ready for use without the need for further sterilization by filtra- tion, which may remove a portion of some compo- nents from the medium. The suitability of such medium, however, will depend on the quality of the water used for the reconstitution of the powder, and it is surprising how little attention is paid to the water supply. The provision of such a sterile and complete powdered medium would be a further step towards the standardization of media because each batch would be pretested, leaving only one variable of the medium (water) in the hands of the individual laboratories. By far the greatest variable component of media is the serum. The exact function of serum is unknown, the species and concentration needed for any particu- TISSUE CULTURE MEDIA AND REAGENTS 749 lar cell culture being determined empirically, but it is clear that serum is essential for almost all cell cultures and for these no substitute for the initiation and active growth of a cell population has been found so far. Serum has at least two functions: it assists in the correct attachment of the cell to a surface and thereafter it plays a part in the metabo- lism of the cell. However, the amino acid composi- tion of the serum does not appear to be important. A small dialysable component is essential and there may be many substances, especially in sera that have been badly prepared, that may be toxic and inhibi- tory. The Cell Culture Committee of the Interna- tional Association of Biological Standardization (1967) suggested 6 years ago that there was an urgent need for the standardization of fetal calf serum and listed 7 measurements that could be made-i.e., sterility, degree of haemolysis, protein content, spec- trographic analysis, presence of free fatty acids, presence of antibiotics, and gamma globulin content. Although these views are widely known, no supplier or user of serum has applied these tests as a routine control. Indeed, 2 years later, it was agreed that the only valid tests of the adequacy of serum are those to determine its ability to support cell growth and its freedom from toxic products. These tests have been applied over the last 5 years and have proved worth while. The incidence of undetected contamination of serum marketed as a sterile product has gained much attention during the last 3 years. The isolation of infectious bovine rhinotracheitis, bovine viral diar- rhoea, and parainfluenza 3 viruses (Kniazeff, 1968; Molander et al., 1972), as well as of mycoplasma (Barile & Kern, 1971) and bacteriophages (Merril et al., 1972; Chu et al., unpublished data, 1973) has not been an infrequent event. It is important that serum marketed as a sterile, nontoxic product should be subjected to appropriate tests to substantiate this claim. The appropriate way to do this appears to be sterilization in the final ampoule by irradiation or chemical sterilization with substances such as 2- oxetanone1 or ethaneperoxoic (" peracetic ") acid. The supply of sterile serum requires further investi- gation. Changes in the laws relating to the slaughter of animals are creating a shortage of calf serum in many countries. There is a pressing need, therefore, to find a substitute for serum. This is not an easy problem, but the use of hormones (insulin) in addition to 1 "f-propiolactone ". serum allows the quantity of serum to be markedly reduced and is, therefore, a step in the right direc- tion. The use of HEPES also allows the concentra- tion of serum to be decreased. It will be many years before there is likely to be a replacement for serum or even a standardized serum fraction, but work to- wards this goal would greatly hasten the standardiza- tion of culture media. CULTURE VESSELS The substance and geometry of culture vessels are most important factors in the growth of uniform cell cultures. For many years it has been known that the volume of medium and the air space above the medium in relation to the surface area of the cell sheet have a marked effect upon the growth of cell cultures. Furthermore, the ability of the same cells to grow in plastic vessels can differ enormously, not only because of the chemical composition of the plastic but also because of the physical nature of its surface. Surface tension at the liquid/air interface and the ability of the surface to be wetted are significant factors in this respect. Many of the details of the production of a satisfactory plastic vessel for the growth of cell cultures are manufacturing secrets, but for each system of cell cultivation it is necessary to specify the source and type of plastic vessel. In considering culture vessels, it should not be forgotten that some kinds of rubber used for the manufacture of stoppers have been shown to be toxic. The simple procedures of washing and sterilizing the vessels must not be overlooked. Cells attach poorly to Pyrex glass washed in hydrochloric acid, but attach well when the glass is rinsed in sodium hydrogen carbonate or sodium hydroxide. Hot-air sterilization is better than steam sterilization, espe- cially if precautions are not taken with the quality of boiler steam, which is often contaminated with light oils. In conclusion, it may be said that, although much progress has been made in the last 10 years towards the standardization of cell culture systems, more could be done. Media made in powdered form in large batches are to be recommended, and sera known to be free from all detectable contaminants and pretested to ensure their suitability for the growth of specific cell cultures are essential. Methods of growing cell cultures under particular condi- tions should be specified in every respect, including details of the culture vessels. Attention to such details would greatly assist progress towards the standardization of cell culture methods in all labora- tories throughout the world. 750 STANDARDIZATION RESUMt LA STANDARDISATION DES MILIEUX ET DES REACTIFS DESTINES AUX CULTURES TISSULAIRES Nos connaissances relatives a la replication des cellules in vitro se sont fortement d6velopp6es durant ces dix dernieres ann&es. On se rend compte toujours davantage de la necessite de standardiser les nombreux composants de ces systemes complexes, mais il reste encore beaucoup ai faire. Un point important a ete acquis lorsqu'on a r6alise que les tissus primaires obtenus a partir d'animaux vivant a l'etat sauvage dans un milieu fortement infecte etaient potentiellement contamines. Un premier progres a consiste en l'emploi de tissus preleves sur des animaux eleves en captivite dans un environnement optimal, mais on s'accorde maintenant sur l'urgence d'etablir des banques de cellules offrant toutes les facilites pour les epreuves preliminaires, la standardisation et le contr6le. La production en grande quantite de milieux en poudre, en assurant une meilleure regularite de la compo- sition, a represente un apport appreciable en vue de la standardisation des nutriments essentiels. La necessite de surveiller le pH pendant toute la dur6e du cycle de croissance a ete reconnue. Le probleme le plus ardu a resoudre reste cependant celui du serum 'a incorporer aux milieux. Ce composant simple, mais de premiere importance, est loin d'etre standardise. On 6tait certes conscient de l'interet de disposer de lots de serum capables d'assurer une croissance satisfaisante, mais ce n'est que recemment que le risque de contamination occulte de produits mis sur le marche a retenu l'attention. On a decele des virus, des mycoplasmes et des bacterio- phages dans des lots consideres comme steriles. La solu- tion semble resider dans la sterilisation du serum dans le recipient final par irradiation ou traitement chimique ou mieux encore dans la mise au point d'un produit de remplacement. L'importance de facteurs accessoires, comme la matiere et la forme des recipients destines aux cultures et les procedes de nettoyage et de sterilisation, ne doit pas etre sous-estimee. De tels details peuvent avoir une influence sur les modalites de croissance des cultures cellulaires. REFERENCES Barile, M. F. & Kern, J. (1971) Proc. Soc. exp. Biol., 138, 432 Ceccarini, C. & Eagle, H. (1971) In: Minutes of the Seventh Meeting of the Cell Culture Committee, Ge- neva, Institute of Hygiene, p. 103 Colombani et al. (1967) Leucocyte and platelet antigens defined by platelet complement fixation test in histo- compatibility testing, Copenhagen, Munksgaard, p. 413 Hayflick, L. & Moorhead, P. S. (1961) Exp. Cell. Res., 25, 585 Hayflick, L. et al. (1964) Nature, 204, 146 Hull, R. N. (1968) The simian viruses, New York, Springer (Virology Monographs No. 2) International Association of Biological Standardization, Cell Culture Committee (1967) Minutes of the Fourth Meeting, Geneva, Institute of Hygiene, p. 19 Jacobs, J. P. (1966) Nature, 210, 100 Jacobs, J. P. & Perkins, F. T. (1969) Bull. Wld Hlth Org., 40, 476 Jacobs, J. P. et al. (1970) Nature, 227, 168 Kniazeff, A. J. (1968) Nat. Cancer Inst. Monogr., 29, 123 Lapeyre, D. et al. (1971) In: Minutes of the Seventh Meeting of the Cell Culture Committee, Geneva, Insti- tute of Hygiene, p. 7 McGarrity, G. J. & Corriell, L. L. (1971) In: Minutes of the Seventh Meeting of the Cell Culture Committee, Geneva, Institute of Hygiene, p. 23 Merril, C. R. et al. (1972) In vitro, 8, 91 Molander, B. et al. (1972) In vitro, 7, 168 Paton, G. R. et al. (1965) Nature, 207, 43 Rose, N. R. et al. (1972) Proc. Soc. exp. Biol., 140, 391 Schulman et al. (1964) Progr. Hemat., 4, 222 Wallace, R. E. et al. (1972) Progr. immunobiol. Standard., 5, 181