- Email: [email protected]
Increased susceptibility to SV40 transformation with development and in vitro aging
EXPERIMENTAL
CELL
RESEARCH
189,222-226
(1990)
Increased Susceptibility to SV40 Transformation with Development and in Vitro Aging TAKAHIRO
KUNISADA,’
DAVID
B. DANNER?
VARDA
FRIEDMAN,
AND EDWARD
L. SCHNEIDER
Laboratory of Molecular Genetics, National Institute on Aging, 4940 Eastern Avenue, Bdtimore, Maryland 21224
The incidence of most cancers increases with aging. To examine whether this increased risk might. be related to a higher susceptibility of older cells to neoplastic transformation, we transfected rat flbroblasts aged in vivo and in vitro with origin-defective SV40 DNA and measured the number of transformed foci. Substantial increases in the number of transformed foci were observed in cells from adult rats when compared with those of cells from embryos or weanlings. Much higher numbers of foci were also obtained at late passage, when 68% or more of the in vitro lifespan had been completed, while no foci were produced from cells at early or middle passage. To control for changes with aging in uptake, integration, or expression of exogenous DNA, parallel cultures were transfected with a G418 resistance gene. The number of G418-resistant colonies did not increase with aging and, in fact, decreased in late passage embryonic cell cultures. Therefore, increased susceptibility to SV40 transformation appears to be a feature of development and in vitro aging in rat, cells. 0 1990 Academic Press, Inc.
INTRODUCTION
The relationship between aging and cancer has been the subject of considerable debate [ 11.The importance of this interaction is reflected by the prevalence of human cancers, half of which occur in individuals over the age of 65 [2]. Two explanations have been offered to account for the increased risk of cancer with aging: (1) that oncogenesis requires a lengthy latent period and (2) that aging increases susceptibility to oncogenesis. Several investigators have attempted to discriminate between these two hypotheses. In one series of experiments, chemical carcinogens were applied to young and old skin which were then transplanted onto young hosts [3, 41. An increased frequency of tumors appearing on 1 Current address: Department of Immunopathology, Kumamoto University Medical School, Honjo 2-2-1, Kumamoto, Japan. ’ To whom reprint requests should be addressed. ’ Current address: Andrus Gerontology Center, University of Southern California, Los Angeles, CA 90027. 0014-4827/90 $3.00 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
222
the older transplanted skin was interpreted to indicate an increased susceptibility of older cells to carcinogenesis [3,4]. By contrast, other researchers have found that carcinogens administered to animals of varying age produce tumors in a time-dependent but not an age-dependent manner [5]. Another possible explanation for the increased prevalence of cancers with aging is that the decline in immune function with age [6] could lead to decreased surveillance and/or killing of malignant cells. In this report, rat fibroblasts grown in tissue culture were exposed to origin-defective SV4Q DNA to examine whether an increased susceptibility to neoplastic transformation accompanies cellular aging in uiuo and in vitro. MATERIALS
AND
METHODS
Zrz vitro studies. Plasmid DNA was prepared by the alkaline-denaturing method [7] followed by two cycles of CsCl gradient centrifugation and phenol/chloroform extractions. Cell cultures were transfected by the calcium phosphate method as modified by Chen and Okayama [8]. Recipient cells were inoculated at a subconfluent level at lo6 cells per lo-cm dish 12 to 14 h before transfection. Salmon sperm carrier DNA (37.5 pg) and plasmid DNA (5 pg) were mixed with calcium phosphate solution and the precipitate was added to each dish. After 4 to 6 h of incubation, cells were washed twice with complete medium and incubated for 16 additional h. The transfected cultures were then trypsinized and split in a ratio of 1:4. When the cells reached confluency, the serum concentration was lowered from 10 to 5%. The cultures were fed every 4 to 5 days. To select for pCDneo transfected cells, 300 pg/ml of G418 was added 1 day after cells were split. G418-resistant colonies and SV40 foci were scored 2 weeks after transfection by methanol fixation and crystal violet staining. To calculate the percent lifespan (donor age/lifespan X lOO%), a value of 17 PDLs was used as the lifespan for embryonic fibroblasts from Wistar rats in the Gerontology Research Center colony. The primary cultures of rat embryo fibroblasts (REF) were prepared from whole 13- to 16day embryos as described below for skin cells from older animals, except that collagenase digestion was performed for 8 to 10 h. Approximately 3-4 X lo6 embryonic cells were inoculated into lo-cm tissue culture dishes. Cumulative population doubling level (PDL) was calculated using harvest and inoculation densities as previously described [S]. Each 1:4 split added 2 PDL and each 1:8 split added 3 PDL. Plating efficiencies of REFs were almost 100% and harvest cell number was 6-8 million, indicating that one doubling occurred after plating. In vivo studies. Preparation of plasmid DNAs and transfection were as described for in vitro studies. Fibroblasts from Wistar rata of different ages were obtained from animals reared at the Gerontology Research Center (maximum lifespan about 28 months). Ventral skin
INCREASED
SV40
FOCI
WITH
samples (5 cm’) were minced in complete Dulbecco’s modified Eagle’s medium (DEEM) supplemented with 10% fetal bovine serum (Hyclone), washed twice with the medium by decant&ion, divided into two lo-cm tissue culture dishes filled with 10 ml of complete growth medium containing 200 units/ml collagenase (GIBCO), and kept at 37°C for 20 to 24 h in a CO, incubator. Skim from 20&y embryos was digested for 8 to 10 h. (Cells from IS-day whole embryos were prepared as for in uitro studies.) Cells were then centrifuged, washed with complete medium to remove residual collagenase, resuspended in complete medium, and inoculated into eight lo-cm petri dishes. Typically cells from young animals became confluent after 4 to 5 days. Cells from old animals needed 1 additional day to reach confluency. Collagenasetreated tissues from adult animals contained significant amounts of lymphocytes, macrophages, and other nonfibroblasti, therefore the plating efficiency of these cells was difficult to determine. To obtain an estimate of plating efficiency, l/1000 dilution aliquots of dissociated cells were placed in lo-cm dishes. After 2 weeks, the number of colonies in each of five replicates was between 200 to 1000. We tberefore estimated that the fibroblasts that grew out from animal tissues experienced approximately 3 to 4 cell divisions prior to the first confluency which was designated as PDL 1. The final density of the primary culture was 3 X 106/10-cm dish in cells from both young and old donors. Cell growth curves begun 3 days after the primary plating were not significantly different between cells derived from young and old donors. Thus our primary cultured cells from animals of various ages were comparable in cell division number from explant&ion. The cellular morphology of primary cultures derived from young and old animals was comparable as well. Southern hybridization analysis of SV40 sequences in trunsfected cells. Individual foci were isolated within 3-mm cloning rings, removed from the dishes by treatment with 0.25% trypsin in EDTA, and transferred to lo-cm dishes. (After 10 additional passages, the cells maintained their transformed appearance.) Cells from each focus were harvested and total cellular DNA was extracted by the guanidinium isotbiocyanate procedure [lo]. For Southern blot analysis, 15 gg of DNA from each focus was digested with EcoRl under the conditions of the supplier (New England Biolabs), subjected to electrophoresis on a 1% agarose gel, and transferred by capillary action to a Gene Screen Plus (DuPont) membrane. Hybridization was carried out in 50% formamide, 1 M NaCl, 0.1% SDS, and 10% dextran sulfate. 32P-labeled probe was prepared by labeling the SV40 ori- plasmid using the random hexamer priming method [ll] and a commercial reaction mix (Oligolabeling Kit, Pbarmacia). Autoradiography of washed filters was by exposure to Kodak XAR-2 film using an intensifying screen (DuPont).
RESULTS
AND
DISCUSSION
In these experiments, cultured fibroblasts were transfected with origin-defective SV40 DNA to determine whether an increased susceptibility to neoplastic transformation accompanies cellular aging. Rat cells were selected for studi since sufficient nonimmune cells could be obtained from the same strain at desired ages. In addition to studying cells as a function of the age of the donor, rat cells were examined as a function of the level of in vitro passage, specifically by population doubling level (PDL). Cells at different levels of passage have frequently been used as a model system to explore cellular aging [12]. To obtain neoplastic transformation, fibroblasts were transfected with SV40 origin minus (ori-) DNA [13] that had been cloned into the bacterial plasmid pMK16 (specifically, we used construct 8-16, missing four nucleotides at the unique Bgll site).
NB
0
FIG. 1. Appearance of SV40 ori- induced foci from skin fibroblasts of newborn (NB) and 24-month-old (0) rats. Two tissue culture dishes are shown containing cells stained with crystal violet. Foci appear as small, intensely staining groupsof cells on a more lightly stained background of nonneoplastic cells; foci have a similar appearance on each plate but differ greatly in number.
The focus assay (see Fig. 1) was selected for the detection of transformed cells rather than assays for growth in semi-solid medium or altered morphology. In preliminary experiments with embryo, young, and old rat fibroblasts, focus assays and soft-agar assays produced similar numbers of transformants after SV40 ori- transfection; in primary cell cultures, the cells were too heterogeneous to permit assays based on morphology. An initial concern in pursuing these studies was that cells of different agesmight have different rates of transfectability with exogenous DNA. For example, young and old cells might differ in their rates of plasmid uptake, integration, or expression of SV40 ori- DNA, thus altering their rates of neoplastic transformation. To examine these potential differences, we transfected parallel cultures with the plasmid pCDneo [8], which when successfully integrated confers stable resistance to the antibiotic G418, a neomycin derivative. The use of a second plasmid controls for age-related differences in DNA uptake and integration; since the neomycin resistance gene in pCDneo is regulated by the same SV40 enhancer and promoter sequences as is the T antigen gene in the SV40 ori- construct, age-related effects on gene transcription are controlled for as well. Embryo fibroblast cultures passaged in vitro displayed substantial increases in their ability to form SV40 foci as they neared the end of their lifespan in tissue culture (Table l), which was approximately 17 PDL. No foci were seen until 11 PDL (approximately 68% of the culture lifespan) and an increase to 24 or 97 foci per lo6 cells was observed between 14 and 15 PDL (approximately 90% of the culture lifespan). The lack of foci in the nine cultures examined before passage 11 indicates that fewer than one focus per lo6 cells occurred in these early passagecultures. Thus, despite the decreased replicative abilities of the late passage cell cultures, they displayed significant ability to be transformed by SV40 ori-
224
KUNISADA TABLE Transformation
In vitro age
Rates per lo6 Cells Aged in Vitro Percent lifespan completed
(population doublings)
1
Rat embryo 1.8 3.0 4.7 6.7 8.6 14.3
fibroblasts-line
Rat embryo 1.0 2.7 5.5 8.9 11.6 14.9
fibroblasts-line
sv40 ori- foci
G418 colonies
1 11 18 28 40 51 84
0 0 0 0 0 24
262 294 363 340 281 2
0 0 0 0 7 97
531 483 856 1440 9 9
2 6 16 32 52 68 88
DNA. Parallel cultures transfected with pCDneo had diminished numbers of G418-resistant colonies with aging, indicating that the changes observed with SV40 ori- were not due to increased rates of uptake, integration, or expression of exogenous DNA. The reason for the large decrease in G418 colonies at very late passage is probably the decreased division potential of these cells [ 141, which could make most of the resulting G418 colonies too small to be counted. It may be that appropriate numbers of SV40 foci are still seen in late passage cells because SV40 is capable of overcoming their poor growth characteristics [l&20]. Cells from older animals can still divide, and a corresponding fall in G418 colonies is not observed in transfected cells as a function of in uiuo age (Table 2). The number of SV40 ori- induced foci also increased as a function of the age of the cell culture donor (Table 2 and Fig. 1). In this case, most of the increase in SV40induced foci occurred between cell cultures derived from embryonic (or newborn) tissue and those obtained from adult animals (Table 2). Only nine transformed foci were detected in four embryonic and neonatal (2-weekold) rat cell cultures compared with 758 foci in transfected cultures from four 2-month-old cultures. In contrast to these changes in SV40 ori- induced foci, the level of G418 colonies produced in these cultures remained essentially the same. The increased susceptibility to SV40 transformation of cells aged in uiuo might be due to previous in uiuo replication. The in uiuo data would then reflect the same phenomenon seen for cells passaged in vitro, where reaching 11 of a possible 17 population doublings is associated with a rise in the SV40 transformation rate (Table 1). However, fibroblasts from mature animals attain over 14 doublings before crisis and our experiments were carried out at population doubling level 1.
ET AL.
The variability seen in the number of G418 colonies for cells from animals of different ages (Table 2) probably reflects the variability inherent in the calcium phosphate method used to transfect these cells. The efficiency of this method is known to vary depending on such subtle factors as the size of the DNA precipitates in each experiment [8]. Since each batch of cells was transfected with both SV40 ori- and neo in one experiment, a l?wer level of variability is found in the SV40 foci:G418 colony ratio for each age range, and this variability is small compared to the change in foci between embryonic and 2-month-old cells. We have also performed these experiments using the lipofectin reagent (Bethesa Research Laboratories), a liposome-mediated transfection protocol [21]; numbers of G418 colonies and SV40 foci are comparable to those obtained by the calcium phosphate method for both embryonic and 24month-old animals (data not shown). An alternative approach to controlling for differences in DNA uptake and integration with age would have been to co-transfect PCDneo and SV40 ori- DNAs, then select with G418 so that foci could be counted against a background of effectively 100% SV40-positive cells. Unfortunately, the number of doublin,gs required for this initial selection would have been so large that the transfected rat cells would have been close to the end of their in vitro lifespan. Since dramatic differences in rates of SV40 transformation are seen with late passage in culture, this approach would not have yielded interpretable data. On the other hand, such an approach may well be appropriate in the case of TABLE Transformation Age of donor (months) Embryo” Embryo Embryo 0.5 2 2 2 2 3 4 6 7 8 12 22 23 24 24 24
Rates Percent lifespan
’ 2 7 7 7 7 11 14 21 25 29 43 79 82 86 86 86
per
2 lo6
Cells Aged in Viuo SV40
foci
8 0 1 0 224 256 160 118 172 28 197 174 399 228 128 176 203 272 571
LIThese two cell cultures were taken from 13-day whole the other cultures in this table are specifically from skin.
G418 colonies 503 531 442 92 536 276 148 329 240 104 494 161 1026 132 56 62 280 144 476 embryos;
all
INCREASED
SV40
FOCI
WITH
FIG. 2. Autoradiograms produced when DNA (15 pg per lane) from individual transformed foci were analyzed by Southern hybridization for the presence of transfected SV40 sequences. Cloned cells from five foci each from SV40 ori- transformed ‘l-month (lanes YlY5) and 24-month (lanes 01-05) rat skin fibroblasts were so analyzed.
human cells with a far greater division potential (55 versus 17 PDL). Together, these results suggest that increased susceptibility to SV40 transformation may occur during early development. Most of the replication of embryonic fibroblasts probably occurs during embryogenesis and the newborn period, when maximum growth occurs. There are probably fewer divisions occurring during adult life. Therefore, embryonic cell cultures at early and middle in vitro passages probably represent in viuo fibroblasts during embryogenesis and early postnatal development and late passage cells probably represent in uiuo adult fibroblasts. Another concern in these studies was that different growth rates in the young and old cells might make the focus assay inaccurate; for example, foci might be hard to see in an embryonic or newborn background because of the high density of the nontransformed cells. This does not seem to have been a problem in our experiments. First, the appearance of foci and background cells appeared reasonably consistent regardless of donor age (Fig. 1). Second, transfection with a different set of oncogenes, myc and ras [22], enabled us to produce hundreds of foci per lo6 embryonic or newborn cells transfected (not shown). Spontaneous transformation is a feature of rat cell cultures. To demonstrate that spontaneous transformation did not materially contribute to the increase in foci with increasing age, two types of control experiments were performed. First, transfection of cell cultures with carrier (genomic) DNA alone was performed in most of these experiments, but foci were never observed. Second, hybridization of DNA and RNA isolated from individual transformed foci confirmed the presence of SV40 DNA and RNA. Southern hybridization analysis of subcloned foci from five young adult and five old rats is presented in Fig. 2. Positive results were also obtained with dot-blots of RNAs from these cells
DEVELOPMENT
AND
AGING
225
(data not shown). These hybridization data show that SV40 ori- DNA can be used to transfect these rodent cells and obtain DNA integration and expression; the genomic DNA controls, carried out for both in uiuo and in vitro aged cells, show that the foci observed are not due to spontaneous transformation. SV40, a monkey virus, probably does not cause cancer in humans. However, it has been shown to stimulate DNA synthesis in senescent human fibroblasts [23,24] and to immortalize human cells in tissue culture at a low rate [ 15-201. Furthermore, SV40 T-antigen binds to the retinoblastoma gene product [25], a protein known to be needed to stop some cells from becoming cancerous. Finally, SV40 is known to cause tumors in rodents [2630]. Thus growth alterations produced by SV40 transfection of cultured cells are likely to be relevant to our understanding of human cancer. Previous studies with SV40 transformation of human fibroblasts, carried out using whole virus and without G418 controls for levels of transfectability, have produced quite variable results, some investigators observing an increase [31], or no change [17], or even a decrease [32] in transformability as a function of in vitro passage level. Our results indicate that individual cells, separated from the immune system and other in uiuo environmental influences, have a dramatic age-related difference in their sensitivity to transformation by SV40 ori- DNA. This difference is seen for rat cells aged both in uiuo and in vitro, and control experiments indicate that it cannot be explained on the basis of increased rates of DNA transfectability or spontaneous neoplastic transformation. The differences in transformation rates are greatest between embryonic and two-month-old animals and between middle and late passage embryonic cells. The mechanism for this developmental/age-related increase is currently unclear. One interesting possibility is that it may reflect differential usage of the SV40 enhancer with development and in vitro aging. It has been shown that the enhancers for SV40 [33] and Polyoma virus [34] are much less active in embryonic cells than in differentiated derivatives. However, differentiation has been shown to produce a decrease in SV40 enhancer activity in adipocytes in culture [35]. Another difficulty with this explanation is that our G418-resistance construct pCDneo also uses the SV40 enhancer [8]. Therefore, if the differences were due to the effect of development on SV40 enhancer utilization, the number of G418 resistant colonies should have displayed comparable changes. Another possibility is that with in vitro passage additional mutations occur in these fibroblasts that favor SV40 transformation, such as activation of protooncogenes or inactivation of tumor suppressor genes. Our control experiments demonstrate that these cells do not form foci without SV40. However, if they were to become immortalized, due to an primary event that could later
226
KUNISADA
cooperate with SV40 T antigen to form a focus, such a primary event would not have been detected in our system. Immortalization has been shown in other systems to cooperate in the formation of foci. This argument is a reasonable possibility in the case of our in vitro results, where the rat cells at late passage are very close to a spontaneous escape from senescence. However, it is unlikely that a mechanism based on somatic mutation could explain our in uiuo results, which reveal the largest difference in SV40 transformability between newborn and 2-month-old animals. Further studies will be initiated to test the connection between these results and altered SV40 enhancer usage with development. It will also be of interest to examine the effect of other oncogenes on the transformability of young and old cell cultures and to extend this work to human cells. We thank Drs. P. A. Cerutti, P. Berg, N. P. Singh, N. Holbrook, A. Fornace, S. Yuspa, and S. Aaronson for their critical comments and suggestions. We are also grateful to Drs. T. F. Williams, V. Cristofalo, R. Sager, and J. Li for helpful advice concerning the manuscript. SV40 ori- and pCDneo were the generous gifts of P. A. Cerutti and H. Okayama, respectively. This work was supported in part by the John D. and Catherine T. MacArthur Foundation Program on Successful Aging. REFERENCES 1. Schneider, E. L. (1985) in Interrelationship among Aging, Cancer, and Differentiation (Pullman, B., Tso, P., and Schneider, E. L., Eds.), pp. l-6, Reidel, Dordecht, The Netherlands. 2. Brody, J. A., and Everett, D. F. (1985) in Interrelationship among Aging, Cancer, and Differentiation (Pullman, B., T’so, P., and Schneider, E. L., Eds.), pp. 7-14, Reidel, Dordecht, The Netherlands. 3. Ebbesen, P. (1984) Science 183,217-2X 4. Ebbesen, P. (1977) J. N&l. Cancer Inst. 68,1057-1060. 5. Peto, R., et ol. (1975) &it. J. Cancer 32,411-426. 6. Hausman, P., and Weksler, M. E. (1985) in Handbook of the Biology of Aging, (Finch, C. E., and Schneider, E. L., Eds.), 2nd ed., pp. 414-432, Van Nostrand-Reinhold, New York. Received December 27,1989 Revised version received March 21,199O
ET AL. 7.
Birnboim,
H., and Doly, S. (1979) Nucleic Acids Res. 7, 1513-
1523.
Chen, C., and Okayama, H. (1987) Mol. Cell. Biol. 7,2745-2752. 9. Schneider, E. L., and Mitsui, Y. (1976) Proc. Natl. Acad. Sci. USA 73,3584-3589. 10. Davis, L., Dibner, M. D., and Battey, J. F. (1986) Basic Methods in Molecular Biology, pp. 130-135, Elsevier, New York. 11. Feinberg, A. P., and Vogel&in, B. (1984) AnaL Biochem. 132, 8.
6-13.
12. Norwood, T. H., and Smith, J. R. (1985) in Handbook of the Biology of Aging, (Finch, C. E., and Schneider, E. L., Eds.), 2nd ed. pp. 291-321, Van Nostrand-Reinhold, New York. 13. Gluzman, Y., Frisque, R. J., and Sambrook, J. (1980) Cold Spring Harbor Symp. Quunt. Biol. 44,293-300. 14. Hayflick, L., and Moorhead, P. S. (1961) Exp. Cell. Res. 26,585621.
15. 16. 17. 18. 19.
Sack, G. H., Jr. (1981) In Vitro 17,1-19. Sager, R., et al. (1983) Proc. Natl. Acad. Sci. USA 80,7601-7605. Huschtacha, L. I., and Holliday, R. (1983) J. Cell Sci. 63,77-99. Stein, G. H. (1985) J. Cell Phys. 125,36-44. Neufeld, D. S., et al. (1987) Mol. Cell. Biol. 7,2794-2802. 0. M., and Smith, J. R. (1987) Mol. Cell Biol. 7, 20. Pereira-Smith, 1541-1544. 21. Felgner, P. L., et al. PFOC. Natl. Acad Sci. USA 84, 7413-7417 (1987). 22. Land, H., Parada, L. F., and Weinberg, R. A. (1983) Nature (L.ondon) 304,596-602. 23. Gorman, S. D., and Cristofalo, V. J. (1982) J. Cell Bill. 96,21a. 24. Ide, T., et al. (1983) Exp. Cell Res. 143,343-349. 25. DeCaprio, J. A., et al. (1988) Cell 64,275-283. 26. Brinster, R. L., et al. (1984) Cell 37,367-379. 27. Bouchard, L., et al. (1984) Virology 135,53-64. 19,185-200. 28. Walsh, J. W., et al. (1986) hk?UFOSUFg63y 29. Pan, S., Abramczuk, J., and Knowles, B. B. (1987) Znt. J. Cancer 39.722-728. 30. 31. 32. 33. 34. 35.
Rachlin, J., Wollmann, R., and Dohrmann, G. (1988) Lab. Znoest. 68,26-30. Jensen, F. C., Koprowski, H., and Ponten, J. A. (1963) PFOC. Natl. Acad. Sci. USA 60,343-345. Matsumara, T., et al. (1980) Exp. Cell Res. 126,453-457. Gorman, C. M., Rigby, P. W. J., and Lane, D. P. (1985) Cell 42, 519-526. Amati, P. (1985) Cell 43,561-562. Djian, P., Phillips, M., and Green, H. (1988) Genes Deu. 2,12511257.
CELL
RESEARCH
189,222-226
(1990)
Increased Susceptibility to SV40 Transformation with Development and in Vitro Aging TAKAHIRO
KUNISADA,’
DAVID
B. DANNER?
VARDA
FRIEDMAN,
AND EDWARD
L. SCHNEIDER
Laboratory of Molecular Genetics, National Institute on Aging, 4940 Eastern Avenue, Bdtimore, Maryland 21224
The incidence of most cancers increases with aging. To examine whether this increased risk might. be related to a higher susceptibility of older cells to neoplastic transformation, we transfected rat flbroblasts aged in vivo and in vitro with origin-defective SV40 DNA and measured the number of transformed foci. Substantial increases in the number of transformed foci were observed in cells from adult rats when compared with those of cells from embryos or weanlings. Much higher numbers of foci were also obtained at late passage, when 68% or more of the in vitro lifespan had been completed, while no foci were produced from cells at early or middle passage. To control for changes with aging in uptake, integration, or expression of exogenous DNA, parallel cultures were transfected with a G418 resistance gene. The number of G418-resistant colonies did not increase with aging and, in fact, decreased in late passage embryonic cell cultures. Therefore, increased susceptibility to SV40 transformation appears to be a feature of development and in vitro aging in rat, cells. 0 1990 Academic Press, Inc.
INTRODUCTION
The relationship between aging and cancer has been the subject of considerable debate [ 11.The importance of this interaction is reflected by the prevalence of human cancers, half of which occur in individuals over the age of 65 [2]. Two explanations have been offered to account for the increased risk of cancer with aging: (1) that oncogenesis requires a lengthy latent period and (2) that aging increases susceptibility to oncogenesis. Several investigators have attempted to discriminate between these two hypotheses. In one series of experiments, chemical carcinogens were applied to young and old skin which were then transplanted onto young hosts [3, 41. An increased frequency of tumors appearing on 1 Current address: Department of Immunopathology, Kumamoto University Medical School, Honjo 2-2-1, Kumamoto, Japan. ’ To whom reprint requests should be addressed. ’ Current address: Andrus Gerontology Center, University of Southern California, Los Angeles, CA 90027. 0014-4827/90 $3.00 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
222
the older transplanted skin was interpreted to indicate an increased susceptibility of older cells to carcinogenesis [3,4]. By contrast, other researchers have found that carcinogens administered to animals of varying age produce tumors in a time-dependent but not an age-dependent manner [5]. Another possible explanation for the increased prevalence of cancers with aging is that the decline in immune function with age [6] could lead to decreased surveillance and/or killing of malignant cells. In this report, rat fibroblasts grown in tissue culture were exposed to origin-defective SV4Q DNA to examine whether an increased susceptibility to neoplastic transformation accompanies cellular aging in uiuo and in vitro. MATERIALS
AND
METHODS
Zrz vitro studies. Plasmid DNA was prepared by the alkaline-denaturing method [7] followed by two cycles of CsCl gradient centrifugation and phenol/chloroform extractions. Cell cultures were transfected by the calcium phosphate method as modified by Chen and Okayama [8]. Recipient cells were inoculated at a subconfluent level at lo6 cells per lo-cm dish 12 to 14 h before transfection. Salmon sperm carrier DNA (37.5 pg) and plasmid DNA (5 pg) were mixed with calcium phosphate solution and the precipitate was added to each dish. After 4 to 6 h of incubation, cells were washed twice with complete medium and incubated for 16 additional h. The transfected cultures were then trypsinized and split in a ratio of 1:4. When the cells reached confluency, the serum concentration was lowered from 10 to 5%. The cultures were fed every 4 to 5 days. To select for pCDneo transfected cells, 300 pg/ml of G418 was added 1 day after cells were split. G418-resistant colonies and SV40 foci were scored 2 weeks after transfection by methanol fixation and crystal violet staining. To calculate the percent lifespan (donor age/lifespan X lOO%), a value of 17 PDLs was used as the lifespan for embryonic fibroblasts from Wistar rats in the Gerontology Research Center colony. The primary cultures of rat embryo fibroblasts (REF) were prepared from whole 13- to 16day embryos as described below for skin cells from older animals, except that collagenase digestion was performed for 8 to 10 h. Approximately 3-4 X lo6 embryonic cells were inoculated into lo-cm tissue culture dishes. Cumulative population doubling level (PDL) was calculated using harvest and inoculation densities as previously described [S]. Each 1:4 split added 2 PDL and each 1:8 split added 3 PDL. Plating efficiencies of REFs were almost 100% and harvest cell number was 6-8 million, indicating that one doubling occurred after plating. In vivo studies. Preparation of plasmid DNAs and transfection were as described for in vitro studies. Fibroblasts from Wistar rata of different ages were obtained from animals reared at the Gerontology Research Center (maximum lifespan about 28 months). Ventral skin
INCREASED
SV40
FOCI
WITH
samples (5 cm’) were minced in complete Dulbecco’s modified Eagle’s medium (DEEM) supplemented with 10% fetal bovine serum (Hyclone), washed twice with the medium by decant&ion, divided into two lo-cm tissue culture dishes filled with 10 ml of complete growth medium containing 200 units/ml collagenase (GIBCO), and kept at 37°C for 20 to 24 h in a CO, incubator. Skim from 20&y embryos was digested for 8 to 10 h. (Cells from IS-day whole embryos were prepared as for in uitro studies.) Cells were then centrifuged, washed with complete medium to remove residual collagenase, resuspended in complete medium, and inoculated into eight lo-cm petri dishes. Typically cells from young animals became confluent after 4 to 5 days. Cells from old animals needed 1 additional day to reach confluency. Collagenasetreated tissues from adult animals contained significant amounts of lymphocytes, macrophages, and other nonfibroblasti, therefore the plating efficiency of these cells was difficult to determine. To obtain an estimate of plating efficiency, l/1000 dilution aliquots of dissociated cells were placed in lo-cm dishes. After 2 weeks, the number of colonies in each of five replicates was between 200 to 1000. We tberefore estimated that the fibroblasts that grew out from animal tissues experienced approximately 3 to 4 cell divisions prior to the first confluency which was designated as PDL 1. The final density of the primary culture was 3 X 106/10-cm dish in cells from both young and old donors. Cell growth curves begun 3 days after the primary plating were not significantly different between cells derived from young and old donors. Thus our primary cultured cells from animals of various ages were comparable in cell division number from explant&ion. The cellular morphology of primary cultures derived from young and old animals was comparable as well. Southern hybridization analysis of SV40 sequences in trunsfected cells. Individual foci were isolated within 3-mm cloning rings, removed from the dishes by treatment with 0.25% trypsin in EDTA, and transferred to lo-cm dishes. (After 10 additional passages, the cells maintained their transformed appearance.) Cells from each focus were harvested and total cellular DNA was extracted by the guanidinium isotbiocyanate procedure [lo]. For Southern blot analysis, 15 gg of DNA from each focus was digested with EcoRl under the conditions of the supplier (New England Biolabs), subjected to electrophoresis on a 1% agarose gel, and transferred by capillary action to a Gene Screen Plus (DuPont) membrane. Hybridization was carried out in 50% formamide, 1 M NaCl, 0.1% SDS, and 10% dextran sulfate. 32P-labeled probe was prepared by labeling the SV40 ori- plasmid using the random hexamer priming method [ll] and a commercial reaction mix (Oligolabeling Kit, Pbarmacia). Autoradiography of washed filters was by exposure to Kodak XAR-2 film using an intensifying screen (DuPont).
RESULTS
AND
DISCUSSION
In these experiments, cultured fibroblasts were transfected with origin-defective SV40 DNA to determine whether an increased susceptibility to neoplastic transformation accompanies cellular aging. Rat cells were selected for studi since sufficient nonimmune cells could be obtained from the same strain at desired ages. In addition to studying cells as a function of the age of the donor, rat cells were examined as a function of the level of in vitro passage, specifically by population doubling level (PDL). Cells at different levels of passage have frequently been used as a model system to explore cellular aging [12]. To obtain neoplastic transformation, fibroblasts were transfected with SV40 origin minus (ori-) DNA [13] that had been cloned into the bacterial plasmid pMK16 (specifically, we used construct 8-16, missing four nucleotides at the unique Bgll site).
NB
0
FIG. 1. Appearance of SV40 ori- induced foci from skin fibroblasts of newborn (NB) and 24-month-old (0) rats. Two tissue culture dishes are shown containing cells stained with crystal violet. Foci appear as small, intensely staining groupsof cells on a more lightly stained background of nonneoplastic cells; foci have a similar appearance on each plate but differ greatly in number.
The focus assay (see Fig. 1) was selected for the detection of transformed cells rather than assays for growth in semi-solid medium or altered morphology. In preliminary experiments with embryo, young, and old rat fibroblasts, focus assays and soft-agar assays produced similar numbers of transformants after SV40 ori- transfection; in primary cell cultures, the cells were too heterogeneous to permit assays based on morphology. An initial concern in pursuing these studies was that cells of different agesmight have different rates of transfectability with exogenous DNA. For example, young and old cells might differ in their rates of plasmid uptake, integration, or expression of SV40 ori- DNA, thus altering their rates of neoplastic transformation. To examine these potential differences, we transfected parallel cultures with the plasmid pCDneo [8], which when successfully integrated confers stable resistance to the antibiotic G418, a neomycin derivative. The use of a second plasmid controls for age-related differences in DNA uptake and integration; since the neomycin resistance gene in pCDneo is regulated by the same SV40 enhancer and promoter sequences as is the T antigen gene in the SV40 ori- construct, age-related effects on gene transcription are controlled for as well. Embryo fibroblast cultures passaged in vitro displayed substantial increases in their ability to form SV40 foci as they neared the end of their lifespan in tissue culture (Table l), which was approximately 17 PDL. No foci were seen until 11 PDL (approximately 68% of the culture lifespan) and an increase to 24 or 97 foci per lo6 cells was observed between 14 and 15 PDL (approximately 90% of the culture lifespan). The lack of foci in the nine cultures examined before passage 11 indicates that fewer than one focus per lo6 cells occurred in these early passagecultures. Thus, despite the decreased replicative abilities of the late passage cell cultures, they displayed significant ability to be transformed by SV40 ori-
224
KUNISADA TABLE Transformation
In vitro age
Rates per lo6 Cells Aged in Vitro Percent lifespan completed
(population doublings)
1
Rat embryo 1.8 3.0 4.7 6.7 8.6 14.3
fibroblasts-line
Rat embryo 1.0 2.7 5.5 8.9 11.6 14.9
fibroblasts-line
sv40 ori- foci
G418 colonies
1 11 18 28 40 51 84
0 0 0 0 0 24
262 294 363 340 281 2
0 0 0 0 7 97
531 483 856 1440 9 9
2 6 16 32 52 68 88
DNA. Parallel cultures transfected with pCDneo had diminished numbers of G418-resistant colonies with aging, indicating that the changes observed with SV40 ori- were not due to increased rates of uptake, integration, or expression of exogenous DNA. The reason for the large decrease in G418 colonies at very late passage is probably the decreased division potential of these cells [ 141, which could make most of the resulting G418 colonies too small to be counted. It may be that appropriate numbers of SV40 foci are still seen in late passage cells because SV40 is capable of overcoming their poor growth characteristics [l&20]. Cells from older animals can still divide, and a corresponding fall in G418 colonies is not observed in transfected cells as a function of in uiuo age (Table 2). The number of SV40 ori- induced foci also increased as a function of the age of the cell culture donor (Table 2 and Fig. 1). In this case, most of the increase in SV40induced foci occurred between cell cultures derived from embryonic (or newborn) tissue and those obtained from adult animals (Table 2). Only nine transformed foci were detected in four embryonic and neonatal (2-weekold) rat cell cultures compared with 758 foci in transfected cultures from four 2-month-old cultures. In contrast to these changes in SV40 ori- induced foci, the level of G418 colonies produced in these cultures remained essentially the same. The increased susceptibility to SV40 transformation of cells aged in uiuo might be due to previous in uiuo replication. The in uiuo data would then reflect the same phenomenon seen for cells passaged in vitro, where reaching 11 of a possible 17 population doublings is associated with a rise in the SV40 transformation rate (Table 1). However, fibroblasts from mature animals attain over 14 doublings before crisis and our experiments were carried out at population doubling level 1.
ET AL.
The variability seen in the number of G418 colonies for cells from animals of different ages (Table 2) probably reflects the variability inherent in the calcium phosphate method used to transfect these cells. The efficiency of this method is known to vary depending on such subtle factors as the size of the DNA precipitates in each experiment [8]. Since each batch of cells was transfected with both SV40 ori- and neo in one experiment, a l?wer level of variability is found in the SV40 foci:G418 colony ratio for each age range, and this variability is small compared to the change in foci between embryonic and 2-month-old cells. We have also performed these experiments using the lipofectin reagent (Bethesa Research Laboratories), a liposome-mediated transfection protocol [21]; numbers of G418 colonies and SV40 foci are comparable to those obtained by the calcium phosphate method for both embryonic and 24month-old animals (data not shown). An alternative approach to controlling for differences in DNA uptake and integration with age would have been to co-transfect PCDneo and SV40 ori- DNAs, then select with G418 so that foci could be counted against a background of effectively 100% SV40-positive cells. Unfortunately, the number of doublin,gs required for this initial selection would have been so large that the transfected rat cells would have been close to the end of their in vitro lifespan. Since dramatic differences in rates of SV40 transformation are seen with late passage in culture, this approach would not have yielded interpretable data. On the other hand, such an approach may well be appropriate in the case of TABLE Transformation Age of donor (months) Embryo” Embryo Embryo 0.5 2 2 2 2 3 4 6 7 8 12 22 23 24 24 24
Rates Percent lifespan
’ 2 7 7 7 7 11 14 21 25 29 43 79 82 86 86 86
per
2 lo6
Cells Aged in Viuo SV40
foci
8 0 1 0 224 256 160 118 172 28 197 174 399 228 128 176 203 272 571
LIThese two cell cultures were taken from 13-day whole the other cultures in this table are specifically from skin.
G418 colonies 503 531 442 92 536 276 148 329 240 104 494 161 1026 132 56 62 280 144 476 embryos;
all
INCREASED
SV40
FOCI
WITH
FIG. 2. Autoradiograms produced when DNA (15 pg per lane) from individual transformed foci were analyzed by Southern hybridization for the presence of transfected SV40 sequences. Cloned cells from five foci each from SV40 ori- transformed ‘l-month (lanes YlY5) and 24-month (lanes 01-05) rat skin fibroblasts were so analyzed.
human cells with a far greater division potential (55 versus 17 PDL). Together, these results suggest that increased susceptibility to SV40 transformation may occur during early development. Most of the replication of embryonic fibroblasts probably occurs during embryogenesis and the newborn period, when maximum growth occurs. There are probably fewer divisions occurring during adult life. Therefore, embryonic cell cultures at early and middle in vitro passages probably represent in viuo fibroblasts during embryogenesis and early postnatal development and late passage cells probably represent in uiuo adult fibroblasts. Another concern in these studies was that different growth rates in the young and old cells might make the focus assay inaccurate; for example, foci might be hard to see in an embryonic or newborn background because of the high density of the nontransformed cells. This does not seem to have been a problem in our experiments. First, the appearance of foci and background cells appeared reasonably consistent regardless of donor age (Fig. 1). Second, transfection with a different set of oncogenes, myc and ras [22], enabled us to produce hundreds of foci per lo6 embryonic or newborn cells transfected (not shown). Spontaneous transformation is a feature of rat cell cultures. To demonstrate that spontaneous transformation did not materially contribute to the increase in foci with increasing age, two types of control experiments were performed. First, transfection of cell cultures with carrier (genomic) DNA alone was performed in most of these experiments, but foci were never observed. Second, hybridization of DNA and RNA isolated from individual transformed foci confirmed the presence of SV40 DNA and RNA. Southern hybridization analysis of subcloned foci from five young adult and five old rats is presented in Fig. 2. Positive results were also obtained with dot-blots of RNAs from these cells
DEVELOPMENT
AND
AGING
225
(data not shown). These hybridization data show that SV40 ori- DNA can be used to transfect these rodent cells and obtain DNA integration and expression; the genomic DNA controls, carried out for both in uiuo and in vitro aged cells, show that the foci observed are not due to spontaneous transformation. SV40, a monkey virus, probably does not cause cancer in humans. However, it has been shown to stimulate DNA synthesis in senescent human fibroblasts [23,24] and to immortalize human cells in tissue culture at a low rate [ 15-201. Furthermore, SV40 T-antigen binds to the retinoblastoma gene product [25], a protein known to be needed to stop some cells from becoming cancerous. Finally, SV40 is known to cause tumors in rodents [2630]. Thus growth alterations produced by SV40 transfection of cultured cells are likely to be relevant to our understanding of human cancer. Previous studies with SV40 transformation of human fibroblasts, carried out using whole virus and without G418 controls for levels of transfectability, have produced quite variable results, some investigators observing an increase [31], or no change [17], or even a decrease [32] in transformability as a function of in vitro passage level. Our results indicate that individual cells, separated from the immune system and other in uiuo environmental influences, have a dramatic age-related difference in their sensitivity to transformation by SV40 ori- DNA. This difference is seen for rat cells aged both in uiuo and in vitro, and control experiments indicate that it cannot be explained on the basis of increased rates of DNA transfectability or spontaneous neoplastic transformation. The differences in transformation rates are greatest between embryonic and two-month-old animals and between middle and late passage embryonic cells. The mechanism for this developmental/age-related increase is currently unclear. One interesting possibility is that it may reflect differential usage of the SV40 enhancer with development and in vitro aging. It has been shown that the enhancers for SV40 [33] and Polyoma virus [34] are much less active in embryonic cells than in differentiated derivatives. However, differentiation has been shown to produce a decrease in SV40 enhancer activity in adipocytes in culture [35]. Another difficulty with this explanation is that our G418-resistance construct pCDneo also uses the SV40 enhancer [8]. Therefore, if the differences were due to the effect of development on SV40 enhancer utilization, the number of G418 resistant colonies should have displayed comparable changes. Another possibility is that with in vitro passage additional mutations occur in these fibroblasts that favor SV40 transformation, such as activation of protooncogenes or inactivation of tumor suppressor genes. Our control experiments demonstrate that these cells do not form foci without SV40. However, if they were to become immortalized, due to an primary event that could later
226
KUNISADA
cooperate with SV40 T antigen to form a focus, such a primary event would not have been detected in our system. Immortalization has been shown in other systems to cooperate in the formation of foci. This argument is a reasonable possibility in the case of our in vitro results, where the rat cells at late passage are very close to a spontaneous escape from senescence. However, it is unlikely that a mechanism based on somatic mutation could explain our in uiuo results, which reveal the largest difference in SV40 transformability between newborn and 2-month-old animals. Further studies will be initiated to test the connection between these results and altered SV40 enhancer usage with development. It will also be of interest to examine the effect of other oncogenes on the transformability of young and old cell cultures and to extend this work to human cells. We thank Drs. P. A. Cerutti, P. Berg, N. P. Singh, N. Holbrook, A. Fornace, S. Yuspa, and S. Aaronson for their critical comments and suggestions. We are also grateful to Drs. T. F. Williams, V. Cristofalo, R. Sager, and J. Li for helpful advice concerning the manuscript. SV40 ori- and pCDneo were the generous gifts of P. A. Cerutti and H. Okayama, respectively. This work was supported in part by the John D. and Catherine T. MacArthur Foundation Program on Successful Aging. REFERENCES 1. Schneider, E. L. (1985) in Interrelationship among Aging, Cancer, and Differentiation (Pullman, B., Tso, P., and Schneider, E. L., Eds.), pp. l-6, Reidel, Dordecht, The Netherlands. 2. Brody, J. A., and Everett, D. F. (1985) in Interrelationship among Aging, Cancer, and Differentiation (Pullman, B., T’so, P., and Schneider, E. L., Eds.), pp. 7-14, Reidel, Dordecht, The Netherlands. 3. Ebbesen, P. (1984) Science 183,217-2X 4. Ebbesen, P. (1977) J. N&l. Cancer Inst. 68,1057-1060. 5. Peto, R., et ol. (1975) &it. J. Cancer 32,411-426. 6. Hausman, P., and Weksler, M. E. (1985) in Handbook of the Biology of Aging, (Finch, C. E., and Schneider, E. L., Eds.), 2nd ed., pp. 414-432, Van Nostrand-Reinhold, New York. Received December 27,1989 Revised version received March 21,199O
ET AL. 7.
Birnboim,
H., and Doly, S. (1979) Nucleic Acids Res. 7, 1513-
1523.
Chen, C., and Okayama, H. (1987) Mol. Cell. Biol. 7,2745-2752. 9. Schneider, E. L., and Mitsui, Y. (1976) Proc. Natl. Acad. Sci. USA 73,3584-3589. 10. Davis, L., Dibner, M. D., and Battey, J. F. (1986) Basic Methods in Molecular Biology, pp. 130-135, Elsevier, New York. 11. Feinberg, A. P., and Vogel&in, B. (1984) AnaL Biochem. 132, 8.
6-13.
12. Norwood, T. H., and Smith, J. R. (1985) in Handbook of the Biology of Aging, (Finch, C. E., and Schneider, E. L., Eds.), 2nd ed. pp. 291-321, Van Nostrand-Reinhold, New York. 13. Gluzman, Y., Frisque, R. J., and Sambrook, J. (1980) Cold Spring Harbor Symp. Quunt. Biol. 44,293-300. 14. Hayflick, L., and Moorhead, P. S. (1961) Exp. Cell. Res. 26,585621.
15. 16. 17. 18. 19.
Sack, G. H., Jr. (1981) In Vitro 17,1-19. Sager, R., et al. (1983) Proc. Natl. Acad. Sci. USA 80,7601-7605. Huschtacha, L. I., and Holliday, R. (1983) J. Cell Sci. 63,77-99. Stein, G. H. (1985) J. Cell Phys. 125,36-44. Neufeld, D. S., et al. (1987) Mol. Cell. Biol. 7,2794-2802. 0. M., and Smith, J. R. (1987) Mol. Cell Biol. 7, 20. Pereira-Smith, 1541-1544. 21. Felgner, P. L., et al. PFOC. Natl. Acad Sci. USA 84, 7413-7417 (1987). 22. Land, H., Parada, L. F., and Weinberg, R. A. (1983) Nature (L.ondon) 304,596-602. 23. Gorman, S. D., and Cristofalo, V. J. (1982) J. Cell Bill. 96,21a. 24. Ide, T., et al. (1983) Exp. Cell Res. 143,343-349. 25. DeCaprio, J. A., et al. (1988) Cell 64,275-283. 26. Brinster, R. L., et al. (1984) Cell 37,367-379. 27. Bouchard, L., et al. (1984) Virology 135,53-64. 19,185-200. 28. Walsh, J. W., et al. (1986) hk?UFOSUFg63y 29. Pan, S., Abramczuk, J., and Knowles, B. B. (1987) Znt. J. Cancer 39.722-728. 30. 31. 32. 33. 34. 35.
Rachlin, J., Wollmann, R., and Dohrmann, G. (1988) Lab. Znoest. 68,26-30. Jensen, F. C., Koprowski, H., and Ponten, J. A. (1963) PFOC. Natl. Acad. Sci. USA 60,343-345. Matsumara, T., et al. (1980) Exp. Cell Res. 126,453-457. Gorman, C. M., Rigby, P. W. J., and Lane, D. P. (1985) Cell 42, 519-526. Amati, P. (1985) Cell 43,561-562. Djian, P., Phillips, M., and Green, H. (1988) Genes Deu. 2,12511257.