ADAPTATION OF HUMAN DIPLOID FIBROBLASTS IN VITRO TO SERUM FROM DIFFERENT SOURCES

Transcription

1 J. Cell Sci. 61, (1983) 289 Printed in Great Britain The Company of Biologists Limited 1983 ADAPTATION OF HUMAN DIPLOID FIBROBLASTS IN VITRO TO SERUM FROM DIFFERENT SOURCES GLEN B. ZAMANSKY*, CARLA ARUNDEL, HATSUMI NAGASAWA AND JOHN B. LITTLE Laboratory of Radiobiology, Harvard University School of Public Health, 665 Huntington Avenue, Boston, Massachusetts 02115, U.SA. SUMMARY The growth of two human diploid skin fibroblast cell lines, originally grown in medium supplemented with foetal bovine serum and later adapted to medium supplemented with newborn bovine, bovine calf or horse serum, has been studied. Prolonged generation times, increased cell volumes and decreased plating efficiencies were observed in cultures grown in newborn bovine, bovine calf or horse serum. In general, the deleterious effects were most severe as a result of growth in bovine calf or horse serum. In the light of the present findings, we believe investigators should exert great caution in switching human fibroblast cultures from foetal bovine serum to alternative sera, even at times of scarcity and high prices. INTRODUCTION Although other serum types have been successfully utilized, the majority of investigators currently use foetal bovine serum as a growth medium supplement for human fibroblast cultures. However, during the past few years, the supply and cost of foetal bovine serum have been extremely variable. At a time of scarcity and high prices, we began to search for alternative serum types. Since our laboratory grows human skin fibroblasts from normal controls and patients with a broad range of genetic disorders, it was important to determine that growth in alternative sera would not influence results of experiments using cells that were originally grown in foetal bovine serum. We therefore examined several parameters of cell growth throughout the life span of a normal control fibroblast cell line and a cell line derived from a patient with hereditary retinoblastoma. These cells were grown in medium supplemented with foetal bovine serum (FBS), newborn bovine serum (NBS), bovine calf serum (BS) or horse serum (HS). MATERIALS AND METHODS Cell lines Human skin fibroblasts were utilized in this study. AG1522 cells, derived from a normal, 3-dayold male were obtained from the Institute for Medical Research (Camden, New Jersey). RbH2BO cells, derived from a skin biopsy of a 1-year-old male with hereditary retinoblastoma, were Author for correspondence at: Boston University School of Medicine, Department of Microbiology, 80 East Concord Street, Boston, Massachusetts 02118, U.S.A.

2 290 G. B. Zamansky and others established in our laboratory by techniques that have been published (Weichselbaum, Epstein & Little, 1976). Medium Cells were grown in Eagle's minimal essential medium (GIBCO) supplemented with 0-9g/l D-glucose, 0-66 mg/1 sodium pyruvate, 25/zg/ml Gentamycin (Schering) and 10% serum (complete medium). Serum types included foetal bovine serum (FBS, GIBCO), newborn bovine serum (NBS, Microbiological Associates), bovine calf serum (BS, Microbiological Associates) and horse serum (HS, Microbiological Associates). All sera were heat-inactivated at 56 C for 30min. Culture conditions AG1522 cells frozen at passage 5 and RbH2BO cells frozen at passage 2 were utilized. Cells that had been stored in liquid nitrogen were thawed and incubated in 25 cm 2 flasks containing 7 ml of complete medium supplemented with FBS. Cultures were maintained at 37 C in a humidified atmosphere of 5 % CO2 and 95 % air. In order to generate sufficient quantities of cells for our experiments, confluent cultures were passed once or twice at a 1:4 dilution in medium containing 10% FBS. When the cultures were again confluent, the medium was removed and the cells were incubated for 2-3 days in complete medium containing FBS, NBS, BS or HS. Thereafter, the cells were grown in 75 cm 2 flasks containing 12 ml of medium supplemented with the appropriate sera and allowed to reach confluence prior to being subcultured at 1:4 dilutions. At early passages, cells were passaged every 6-8 days. At higher passage levels, additional time was required between passes. Culture medium was replaced at 4-5-day intervals and within 48 h prior to experiments. Since the cells were subcultured at 1:4 dilutions, each passage represents two mean population doublings. Subculture conditions Cultures were washed once with calcium- and magnesium-free Earle's balanced salt solution (EBSS, GIBCO). The cells were then exposed to 0-25 % trypsin in EBSS for approximately 2min at 37 C and subsequently resuspended in medium containing the appropriate serum. Growth kinetics and cell volume Cells from confluent stock cultures that had been subcultured at a 1: 4 dilution were detached with 0-25 % trypsin and resuspended in complete medium. A total of 5 X 10 4 cells in 5 ml of medium were plated in 25 cm 3 flasks. Duplicate cultures were trypsinized and counted for a 10-day period. Generation times were calculated from the exponential portions of the growth curves plotted from these counts. Volumes of trypsinized cells from confluent cultures were obtained using a Coulter counter, which was calibrated with paper mulberry pollen as a standard. Plating efficiencies Cells from confluent stock cultures that had been passaged at a 1: 4 dilution were detached with 0-25 % trypsin and resuspended in complete medium. Cell suspensions were diluted appropriately in order to inoculate 100 mm Petri dishes (containing 10 ml of complete medium) with 250 AG1522 cells or 400 RbH2BO cells. One week later the culture medium was replaced with fresh medium. Approximately 2 weeks after plating, the cultures were rinsed twice with EBSS and fixed with 10 % buffered formalin. Cultures were then stained with 10 % Giemsa stain and colonies containing more than 50 cells were scored. Plating efficiencies were calculated by dividing the number of colonies by the number of cells plated. Three plates were counted in determining each experimental value. Fig. 1. Generation times of AG1522 (A) and RbH2BO (B) cells grown in medium supplemented with: FBS (O O); NBS ( ); BS (A A); or HS ( --D). AG1522 and RbH2BO cells were continuously subcultured in these test sera beginning at passages 8 and 5, respectively. Generation times were determined from the exponential portion of survival curves. Since the cultures were passaged at 1:4 dilutions, each passage represents two mean population doublings.

3 Serum type and human fibroblasts A AGI q t A If a// ^y^,. " V o o" 0 "j i 1 i RbH2BO Passage number Fig

4 292 G. B. Zamansky and others RESULTS During our search for a substitute for FBS, we screened ten lots of NBS, nine lots of BS and two lots of HS received from six companies. Various human cell lines were cultured in the sera for 1-2 weeks and examined for growth, colony-forming ability and cellular appearance. The sera used below were considered the most promising available lots of each type based on this preliminary screening. The FBS utilized was a lot in which we were routinely growing human cells at the time this work started. In order to determine the effect of different sera on cellular growth properties, several parameters were measured. Growth curves were obtained at various passage levels of AG1522 and RbH2BO cells, which had been maintained in medium supplemented with the appropriate test serum beginning at passages 8 and 5, respectively. Generation times determined from these curves are presented in Fig. 1. Since the cells were subcultured at 1:4 dilutions each passage represents two mean population doublings. Cells grown in FBS maintained the shortest doubling times. After eight passages in medium supplemented with NBS, BS or HS, AG1522 cells began to replicate more slowly than those in FBS. By the eleventh subpassage these cultures exhibited generation times approximately twofold higher than cells grown in FBS. RbH2BO cells were considerably more sensitive to growth in NBS, BS and HS. Following only four passages in these sera, the generation times were twice those in FBS, and after eight passages the cells in BS and HS no longer grew sufficiently well to warrant additional subcultures. Cells grown in NBS were subcultured once more. The volumes of cells from confluent stock cultures were also determined. As seen in Fig. 2A, the median volume of AG1522 cells grown in FBS did not change significantly during these experiments. The values for cells grown in NBS, BS or HS were within the range observed for FBS through the first 10 passages in these sera. However, by the eleventh passage in the test sera volumes were greater than those found for FBS cultures. RbH2BO cell volumes remained constant through passage 13 in FBS and gradually increased at later passages (Fig. 2B). Subculturing three to five times in BS or HS was sufficient to result in increased volumes for these cells. As was found for generation times, growth in NBS increased cell volumes, but the values were altered less dramatically than in BS or HS. The plating efficiencies of AG1522 and RbH2BO cells at low cell density are shown in Fig. 3. Plating efficiencies dropped at earlier passages when the cells were grown in NBS, BS or HS instead of FBS. The pattern of change for RbH2BO cells was very much like that observed above for generation times and cell volumes. By four passages in the alternative sera, the plating efficiencies were below those of FBS. However, AG1522 cells grown in NBS or BS demonstrated decreased plating efficiencies at earlier passages than changes had been found in generation times or cell volumes. Although the plating efficiencies of AG1522 cells grown in HS do not appear to fall until several passages later, these cells displayed an unusual colony morphology throughout their passage. As seen in Fig. 4, AG1522 cells grown in FBS formed colonies consisting of fibroblasts in the swirled pattern usually observed for such human cells. Cells cultured in HS, on the other hand, grew in a patchy disoriented

5 Serum type and human fibroblasts 293 AG * f> CM o p o in * * 8 o Passage number Fig. 2. Median volume of AG 1522 (A) and RbH2BO (B) cells in confluent stock cultures. FBS (O O); NBS ( ); BS ( A); HS ( ). The range of values for passage 8 through passage 20 AG1522 cells grown in FCS is indicated by the shaded area.

6 294 G- B. Zamansky and others 10 AG t 001!fe Passage number Fig. 3. Plating efficiencies of AG1522 (A) and RbH2BO (B) cells cultured at low cell density. Symbols are the same as in Figs 1 and 2. Plating efficiencies were calculated by dividing the number of colonies containing at least SO cells by the number of cells plated. Each experimental point represents the average of three plates.

7 Serum type and human fibroblasts 295 Fig. 4. Appearance of AG1522 cells in colonies grown in FBS (A) or HS (B). TWO weeks after plating, the cultures were fixed with 10% formalin and stained. In order to ensure sufficient resolution only a portion of each individual colony was photographed. X100. manner. This was not observed for colonies grown in the other sera nor in RbH2BO colonies grown in HS. DISCUSSION We have observed increased generation times, increased cell volumes and decreased plating efficiencies in two cell lines originally grown in FBS and later cultured in medium supplemented with NBS, BS or HS. In general, the deleterious effects of the substitute sera were more severe in BS and HS than in NBS, and more pronounced at earlier passages in RbH2BO than AG1522 cells. The severity of the changes in RbH2BO cultures was not unique to this cell line since we have observed similar early alterations in cells derived from a patient with xeroderma pigmentosum (unpublished data). These findings may be indicative of a pattern of premature senescence in our cells cultured in NBS, BS or HS. The process of cell aging has been studied in vitro utilizing several parameters, amongst which are generation time, cell size, ability to synthesize DNA and chromosomal stability (Saksela & Moorhead, 1963; Macieira-Coelho, Ponten & Philipson, 1966; Cristofalo & Sharf, 1973; Absher, Absher & Barnes, 1974; Yanishevsky, Mendelsohn, Mayall & Cristofalo, 1974; Bowman, Meek & Daniel, 1975; Matsumura, Pfendt & Hayflick, 1979; Matsumura, Zerrudo & Hayflick, 1979; Schneider & Fowlkes, 1976). Such studies indicate that there is a strong correlation

8 296 G. B. Zamansky and others between senescence, prolonged generation times and increased cell size. The prolonged division times appear to be due to an increase in the average generation time of cycling cells and a progressive increase in the fraction of cells proceeding towards a state of growth arrest in which they are no longer capable of synthesizing DNA (Macieira-Coelhoefa/. 1966; Cristofalo & Sharf, 1973; Absher e* a/. 1974; Bowman etal. 1975; Merz & Ross, 1969). We have obtained preliminary data that indicate that growth of RbH2BO cells in NBS, BS, or HS results in a smaller percentage of cells able to incorporate [ 3 H]thymidine. Thus our cultures also appear to consist of such a changing population of cells. It has been suggested that the shift towards larger cells is due to a greater number of slowly or non-dividing cells (Bowman et al. 1975). Although these cells have a diminished capacity to synthesize DNA, the continued synthesis of other molecules (Macieira-Coelho et al. 1966; Cristofalo & Kritchevsky, 1969) may contribute to the increase in size. In addition to the parameters discussed above, we also investigated the plating efficiency of AG1522 and RbH2BO cells at low density, since many experiments are performed under such conditions in our laboratory. The plating efficiencies were generally found to correlate with the ability of the individual sera to support cell growth. The observation that AG1522 cells grown in HS formed colonies with a peculiar morphology remains unexplained. Stock cultures of these cells formed normal monolayers at early passages in spite of their inability to form normal colonies. From this study it is difficult to make generalizations concerning the effect of other preparations of NBS, BS or HS, since we have systematically investigated only one lot of each serum type. However, it is important to note that prior to selecting the particular lots utilized in this study, we preliminarily screened a total of 19 lots of NBS and BS from six companies. The lot of NBS chosen was considered the best serum in terms of growth, colony-forming ability and cellular morphology. It is also difficult to predict what effect the different sera would have had on cellular growth if our cell lines had originally been established in these sera. It is clear from earlier studies that bovine serum can be used successfully to establish and grow human fibroblasts (Hayflick & Moorhead, 1961; Saksela & Moorhead, 1963). However, most investigators presently use FBS for these purposes. The two major cell banks from which human skin fibroblasts can be purchased in the United States freeze their cell stocks in medium containing FBS, distribute their cultures in FBS and recommend that the cells be grown in FBS. In the light of our findings, we believe investigators should exert great caution in switching human fibroblast cultures from FBS to alternative sera, even at times of scarcity and high prices. This work was supported by grants CA11751 and ES00002 from the United States National Institutes of Health. We would like to thank June Mouradjian for her excellent technical assistance and Dr Ralph Weichselbaum for providing us with RbH2BO cells. REFERENCES ABSHER, P. M., ABSHER, R. G. & BARNES, W. O. (1974). Genealogies of clones of diploid fibroblasts. Cinemicrophotographic observations of cell division patterns in relation to population age. Expl Cell Res. 88,

9 Serum type and human fibroblasts 297 BOWMAN, P. D., MEEK, R. L. & DANIEL, C. W. (1975). Aging of humanfibroblastsin vitro. Expl Cell Res. 93, CRISTOFALO, V. J. & KRITCHEVSKY, D. (1969). Cell size and nucleic acid content in the diploid cell line WI38 during aging. Med. Exp. 19, CRISTOFALO, V. J. &SHARF, B. (1973). Cellular senescence and DNA synthesis. Expl Cell Res. 76, HAYFLICK, L. & MOORHEAD, P. S. (1961). The serial cultivation of human diploid cell strains. Expl Cell Res. 25, MACIEIRA-COELHO, A., PONT N, J. & PHILIPSON, L. (1966). The division cycle and RNA synthesis in diploid human cells at different passage levels in vitro. Expl Cell Res. 42, MATSUMURA, T., PFENDT, E. A. & HAYFLICK, L. (1979). DNA synthesis in the human diploid cell strain WI-38 during in vitro aging: An autoradiography study. /. Geront. 34, MATSUMURA, T., ZERRUDO, Z. & HAYFLICK, L. (1979). Senescent human diploid cells in culture: Survival, DNA synthesis and morphology. J. Geront. 34, MERZ, G. S. & Ross, J. D. (1969). Viability of human diploid cells as a function of in vitro age. J. cell. Physiol. 74, SAKSELA, E. & MOORHEAD, P. S. (1963). Aneuploidy in the degenerative phase of serial cultivation of human cell strains. Proc. natn. Acad. Sci. U.S.A. 50, SCHNEIDER, E. L. & FOWLKES, B. J. (1976). Measurement of DNA content and cell volume in senescent human fibroblasts utilizing flow multiparameter single cell analysis. Expl Cell Res. 98, WEICHSELBAUM, R., EPSTEIN, J. & LITTLE, J. B. (1976). A technique for developing established cell lines from human osteosarcomas. In Vitro 12, YANISHEVSKY, R. M., MENDELSOHN, M., MAYALL, B. H. & CRISTOFALO, V. J. (1974). Proliferative capacity and DNA content of aging human diploid cells in culture: A cytophotometric and autoradiographic analysis. J. cell. Physiol. 84, (Received 11 August 1982)

10