- Email: [email protected]
Molecular genetic studies of cellular senescence
ExperimentalGerontology,Vol.27, pp. 519-522, 1992 Printed in the USA.All rightsreserved.
MOLECULAR
GENETIC
0531-5565/92 $5.00 + .00 Copyright© 1992PergamonPressLtd.
STUDIES
OF CELLULAR
SENESCENCE
O . M . PEREIRA-SMITH t-3 and Y. NING 1'2
tHuffington Center on Aging, 2Division of Molecular Virology, and 3Department of Medicine, Baylor College of Medicine, One Baylor Plaza, Houston, Texas 77030
Abstract - - The limited doubling potential of normal cells in culture was first proposed
as a model for cellular aging by Hayflick in 196 I. This phenomenon of in vitro cellular senescence is now well documented for a number of different normal human cell types. In an attempt to determine whether random events or programmed genetic processes were responsible for cellular aging, we performed a series of cell fusion studies. We determined that hybrids from fusion of normal with immortal human cells had limited proliferative potential, indicating that senescence is a dominant phenotype. We exploited the fact that immortality was recessive to assign a large number of different immortal human cell lines to four complementation groups for indefinite division. More recently, we have determined that the introduction of a single normal human chromosome 4 into HeLa (cervical carcinoma) cells by microcell fusion induced senescence in this immortal line. The results of these whole cell and microcell fusion studies support the hypotheses that propose senescence results from active, genetic mechanisms. Key Words: limited doubling potential, cellular aging, cell fusion, senescence, HeLa
INTRODUCTION IT IS generally considered that aging o f the individual is a highly complex process and the result o f a large variety of changes. T h e latter do not occur in a systematic, predictable m a n n e r in each individual, and this genetic heterogeneity tends to complicate the study o f the aging h u m a n even further. In addition, it is difficult to distinguish changes that are the result o f n o r m a l aging f r o m the disease processes that a c c o m p a n y it. It is for these reasons that gerontologists have sought simpler model systems to study, with the goal o f eventually applying results obtained with the models to the aging individual. T h e model system we have elected to use is that of the normal h u m a n cell in culture. T h e limited proliferative capacity typical of n o r m a l cells was first observed by Hayflick a n d M o o r h e a d in 196 I. Hayflick (1965) proposed this p h e n o m e n o n as a model for aging at the cellular level on the basis that it reflected what occurred in cells capable of replication in vivo. Evidence in support o f the model, described in detail by Drs. Hayflick and Cris-
Correspondence to: O.M. Pereira-Smith. 519
520
O.M. PEREIRA-SMITH AND Y. NING
tofalo, has led to its establishment and use in many laboratories throughout the world (Goldstein, 1990). Two approaches have been taken in determining the mechanisms that lead to cellular senescence (ie., the loss of division potential of the normal cell). One has involved the determination of changes that occur in the normal cell as it proceeds through its in vitro life span. The other has focused on cells that have escaped from senescence, in an attempt to understand what has occurred to allow these cells to proliferate indefinitely in vitro and become immortal. Our studies to understand the basic mechanisms of cellular senescence have utilized both approaches. WHOLE CELL FUSION STUDIES A number of hypotheses have been proposed to explain the phenomenon of cellular senescence, and these fall into two broad groups. One favors the idea that senescence involves genetically based mechanisms and is the result of a genetic program or specific changes in gene expression. The other proposes that senescence is the result of accumulation of random damage. To distinguish these hypotheses, we fused normal with immortal human cells and determined the proliferation of the hybrids. If random damage did cause senescence, one would have to assume the immortal cell would have acquired new mechanisms to deal with the damage. The prediction would be that these new mechanisms would be present in a hybrid, which should therefore exhibit an immortal phenotype.
Normal-immortal fusions We initiated these studies by fusing normal human diploid fibroblasts with immortal Simian Virus 40 (SV40)-transformed cell lines (Pereira-Smith and Smith, 1981). A biochemical selection system was used to select the hybrids. Fusion products were inoculated into selection medium at very low numbers (20-200 cells/60 mm tissue culture dish) and incubated undisturbed for 2 weeks. At this time, large clones of > 2 5 0 cells were located by scanning under a phase contrast microscope and subcultured. The remaining clones were fixed, stained and the number of cells within each clone counted. We found that the majority of the hybrid clones were small and comprised of fewer than 200 cells. Those that could be subcultured achieved a range of population doubling (PD) from 16-42, but all eventually lost division potential and became nondividing. These results indicated that the phenotype of senescence is dominant, and that immortal cells result from recessive changes in genes involved in the process of senescence. They also lent support to the hypotheses that propose cellular senescence results from active, genetic events rather than from random damage. All clones continued to express SV40 T-antigen, even when they had ceased division, indicating that viral genes alone were incapable of inducing proliferation in the hybrids. We extended these studies to include other normal cell types. Additional fibroblast cell lines (Pereira-Smith and Smith, 1983), vascular endothelial cells (unpublished observations), and T lymphocytes (Pereira-Smith et al., 1990a) were used as the normal parent in such fusions, and we obtained the same result. The hybrids had limited life span. We also fused a wide variety of immortal human cells (Pereira-Smith and Smith, 1983; PereiraSmith et aL, 1990b) with the normal cells and again observed that the hybrids exhibited limited life span. These results confirmed a genetic basis for cellular senescence.
MOLECULAR GENETIC STUDIES OF CELLULAR SENESCENCE
52 |
Immortal-immortal cell fusions We exploited the fact that the immortal phenotype was recessive to identify the number of genes or processes involved in the genetic program of cellular senescence. We did this by fusing various immortal cell lines with each other. If the parental cell lines used in the fusion had immortalized as a result of recessive defects in the same gene, the defect would continue to be present in the genotype of the hybrid, which would be immortal. The two cell lines would then be assigned to the same complementation group for indefinite division. However, if the immortal parents fused had escaped from senescence by different genetic changes, these would be complemented in the hybrid, which would exhibit limited life span. In this case, the cell lines would be assigned to different complementation groups for indefinite division. The number of groups identified would indicate the number of genes or processes involved in cellular senescence. To date, we have fused 30 immortal human cell lines with each other and identified four complementation groups (Pereira-Smith and Smith, 1988; unpublished observations). No cell line has assigned to more than one group, indicating that a small number of highly specific genes are involved in senescence. These must be modified for a cell to become immortal. These results are consistent with the facts that spontaneous immortalization of human cells from normal individuals has never been reported, and that the frequency of induced immortalization of human cells is low. We included a wide variety of cell lines in the analysis to determine correlations, if any, with group assignment. We observed no relationship between group assignment and cell type, embryonal layer of origin, type of tumor, or expression of an activated oncogene. The single correlation observed was the assignment of six of seven immortal SV40-transformed cells to the same complementation group (Pereira-Smith and Smith, 1983, 1987). This observation led to the collaborative series of studies, described by Dr. W. Wright, in which we have determined that SV40 large T-antigen expression is required for cell division in cells postimmortalization (Wright et al., 1989), and have identified some of the functions o f t antigen that are involved. In addition, the assignment of immortal cell lines to specific groups allowed us to take a focused approach to identify the chromosomes and eventually the genes involved in cellular senescence.
MICROCELL-MEDIATED CHROMOSOME TRANSFER STUDIES We proceeded with the relatively new technique of microcell fusion, introducing normal single human chromosomes into immortal cell lines representing the various complementation groups, to determine an effect on cell division. We began our studies with chromosome 11. The microcell donor was a monochromosomal mouse-human hybrid that retained a single human derivative chromosome 11 in which most of chromosome 11 had translocated to a small portion of the X chromosome, retaining the hypoxanthine guanine phosphoribosyl transferase (HPRT) gene. We introduced this chromosome into H P R T - cell lines representative of 3 of 4 of the complementation groups (Group C cell lines could not be used because H P R T - subclones were not obtained after many attempts). Chromosome 11 had no effect on the proliferation of the ceils, indicating that genes on this chromosome were not involved in cellular senescence (Ning et al., 1991).
522
O.M.PEREIRA-SMITHANDY. NING
However, when we introduced an intact normal human chromosome 4 into cell lines representing the various complementation groups, senescence was induced in HeLa cells (Group B), but not cell lines assigned to the other groups (unpublished results). Work is in progress to determine whether this effect occurs in other cell lines within this group, and to identify the region of the chromosome that is involved in cellular senescence. CONCLUSIONS In summary, we believe that our data clearly demonstrate the dominance of the senescence phenotype and support the hypothesis that cellular aging is an active, genetically based process rather than the result of accumulation of random damage. The microcell fusion results indicate that it will be feasible to use this approach to identify the genes involved in cellular senescence. Once the genes are identified, their regulation and involvement in senescence of various normal cells, such as vascular endothelial cells and T lymphocytes, can be examined. It may then be possible to outline strategies for intervention in the aging individual to, achieve the goal of gerontologists: improvement in the quality of life of the elderly. Acknowledgments - - This work was supported by National Institutes of Health grants AGO5333 and POIAGO7123, and the Pilgeram Fund for Aging Research.
REFERENCES GOLDSTEIN, S. Replicative senescence: The human fibroblast comes of age. Science 249, 1129-1133, 1990. HAYFLICK, L. The limited in vitro lifetime of human diploid cell strains. Exp. Cell Res. 37, 614-636, 1965. HAYFLICK, L. and MOORHEAD, P.S. The serial cultivation of human diploid strains. Exp. CellRes. 25, 585621, 1961. NING, Y., SHAY, J.W., LOVELL, M., TAYLOR, L., LEDBETTER, D.H., and PEREIRA-SMITH, O.M. Tumor suppression by chromosome 11 is not due to cellular senescence. Exp. CellRes. 192, 220-226, 1991. PEREIRA-SMITH, O.M., ROBETORYE, S., NING., Y., and ORSON, F. Hybrids from fusion of normal human T lymphocytes with immortal human cells exhibit limited lifespan. J. Cell. Physiol. 144, 546-549, 1990a. PEREIRA-SMITH, O.M. and SMITH, J.R. Expression of SV40 T antigen in finite lifespan hybrids of normal SV40 transformed cells. Som. Cell Genet. 7, 411-421, 1981. PERERIA-SMITH, O.M. and SMITH, J.R. Evidence for the recessive nature of cellular immortality. Science 221,964-966, 1983. PEREIRA-SMITH, O.M. and SMITH, J.R. Functional simian virus 40 T antigen is expressed in hybrid cells having finite proliferative potential. Mol. Cell. Biol. 7, 1541-1544, 1987. PEREIRA-SMITH, O.M. and SMITH, J,R. Genetic analysis of indefinite division in human cells: Identification of four complementation groups. Proc. Natl. Acad. Sci. U. S. A. 85, 6042-6046, 1988. PEREIRA-SMITH, O.M., STEIN, G.H., ROBETORYE, S., and MEYER-DEMAREST, S. Immortal phenotype of the HeLa variant D98 is recessive in hybrids formed with normal human fibroblasts..L Cell. Physiol. ' 143,222-225, 1990b. WRIGHT, W.E., PEREIRA-SMITH, O.M., and SHAY, J.W. Reversible cellular senescence: Implications for immortalization of normal human diploid fibroblasts. Mol. Cell. Biol. 9, 3088-3092, 1989.
MOLECULAR
GENETIC
0531-5565/92 $5.00 + .00 Copyright© 1992PergamonPressLtd.
STUDIES
OF CELLULAR
SENESCENCE
O . M . PEREIRA-SMITH t-3 and Y. NING 1'2
tHuffington Center on Aging, 2Division of Molecular Virology, and 3Department of Medicine, Baylor College of Medicine, One Baylor Plaza, Houston, Texas 77030
Abstract - - The limited doubling potential of normal cells in culture was first proposed
as a model for cellular aging by Hayflick in 196 I. This phenomenon of in vitro cellular senescence is now well documented for a number of different normal human cell types. In an attempt to determine whether random events or programmed genetic processes were responsible for cellular aging, we performed a series of cell fusion studies. We determined that hybrids from fusion of normal with immortal human cells had limited proliferative potential, indicating that senescence is a dominant phenotype. We exploited the fact that immortality was recessive to assign a large number of different immortal human cell lines to four complementation groups for indefinite division. More recently, we have determined that the introduction of a single normal human chromosome 4 into HeLa (cervical carcinoma) cells by microcell fusion induced senescence in this immortal line. The results of these whole cell and microcell fusion studies support the hypotheses that propose senescence results from active, genetic mechanisms. Key Words: limited doubling potential, cellular aging, cell fusion, senescence, HeLa
INTRODUCTION IT IS generally considered that aging o f the individual is a highly complex process and the result o f a large variety of changes. T h e latter do not occur in a systematic, predictable m a n n e r in each individual, and this genetic heterogeneity tends to complicate the study o f the aging h u m a n even further. In addition, it is difficult to distinguish changes that are the result o f n o r m a l aging f r o m the disease processes that a c c o m p a n y it. It is for these reasons that gerontologists have sought simpler model systems to study, with the goal o f eventually applying results obtained with the models to the aging individual. T h e model system we have elected to use is that of the normal h u m a n cell in culture. T h e limited proliferative capacity typical of n o r m a l cells was first observed by Hayflick a n d M o o r h e a d in 196 I. Hayflick (1965) proposed this p h e n o m e n o n as a model for aging at the cellular level on the basis that it reflected what occurred in cells capable of replication in vivo. Evidence in support o f the model, described in detail by Drs. Hayflick and Cris-
Correspondence to: O.M. Pereira-Smith. 519
520
O.M. PEREIRA-SMITH AND Y. NING
tofalo, has led to its establishment and use in many laboratories throughout the world (Goldstein, 1990). Two approaches have been taken in determining the mechanisms that lead to cellular senescence (ie., the loss of division potential of the normal cell). One has involved the determination of changes that occur in the normal cell as it proceeds through its in vitro life span. The other has focused on cells that have escaped from senescence, in an attempt to understand what has occurred to allow these cells to proliferate indefinitely in vitro and become immortal. Our studies to understand the basic mechanisms of cellular senescence have utilized both approaches. WHOLE CELL FUSION STUDIES A number of hypotheses have been proposed to explain the phenomenon of cellular senescence, and these fall into two broad groups. One favors the idea that senescence involves genetically based mechanisms and is the result of a genetic program or specific changes in gene expression. The other proposes that senescence is the result of accumulation of random damage. To distinguish these hypotheses, we fused normal with immortal human cells and determined the proliferation of the hybrids. If random damage did cause senescence, one would have to assume the immortal cell would have acquired new mechanisms to deal with the damage. The prediction would be that these new mechanisms would be present in a hybrid, which should therefore exhibit an immortal phenotype.
Normal-immortal fusions We initiated these studies by fusing normal human diploid fibroblasts with immortal Simian Virus 40 (SV40)-transformed cell lines (Pereira-Smith and Smith, 1981). A biochemical selection system was used to select the hybrids. Fusion products were inoculated into selection medium at very low numbers (20-200 cells/60 mm tissue culture dish) and incubated undisturbed for 2 weeks. At this time, large clones of > 2 5 0 cells were located by scanning under a phase contrast microscope and subcultured. The remaining clones were fixed, stained and the number of cells within each clone counted. We found that the majority of the hybrid clones were small and comprised of fewer than 200 cells. Those that could be subcultured achieved a range of population doubling (PD) from 16-42, but all eventually lost division potential and became nondividing. These results indicated that the phenotype of senescence is dominant, and that immortal cells result from recessive changes in genes involved in the process of senescence. They also lent support to the hypotheses that propose cellular senescence results from active, genetic events rather than from random damage. All clones continued to express SV40 T-antigen, even when they had ceased division, indicating that viral genes alone were incapable of inducing proliferation in the hybrids. We extended these studies to include other normal cell types. Additional fibroblast cell lines (Pereira-Smith and Smith, 1983), vascular endothelial cells (unpublished observations), and T lymphocytes (Pereira-Smith et al., 1990a) were used as the normal parent in such fusions, and we obtained the same result. The hybrids had limited life span. We also fused a wide variety of immortal human cells (Pereira-Smith and Smith, 1983; PereiraSmith et aL, 1990b) with the normal cells and again observed that the hybrids exhibited limited life span. These results confirmed a genetic basis for cellular senescence.
MOLECULAR GENETIC STUDIES OF CELLULAR SENESCENCE
52 |
Immortal-immortal cell fusions We exploited the fact that the immortal phenotype was recessive to identify the number of genes or processes involved in the genetic program of cellular senescence. We did this by fusing various immortal cell lines with each other. If the parental cell lines used in the fusion had immortalized as a result of recessive defects in the same gene, the defect would continue to be present in the genotype of the hybrid, which would be immortal. The two cell lines would then be assigned to the same complementation group for indefinite division. However, if the immortal parents fused had escaped from senescence by different genetic changes, these would be complemented in the hybrid, which would exhibit limited life span. In this case, the cell lines would be assigned to different complementation groups for indefinite division. The number of groups identified would indicate the number of genes or processes involved in cellular senescence. To date, we have fused 30 immortal human cell lines with each other and identified four complementation groups (Pereira-Smith and Smith, 1988; unpublished observations). No cell line has assigned to more than one group, indicating that a small number of highly specific genes are involved in senescence. These must be modified for a cell to become immortal. These results are consistent with the facts that spontaneous immortalization of human cells from normal individuals has never been reported, and that the frequency of induced immortalization of human cells is low. We included a wide variety of cell lines in the analysis to determine correlations, if any, with group assignment. We observed no relationship between group assignment and cell type, embryonal layer of origin, type of tumor, or expression of an activated oncogene. The single correlation observed was the assignment of six of seven immortal SV40-transformed cells to the same complementation group (Pereira-Smith and Smith, 1983, 1987). This observation led to the collaborative series of studies, described by Dr. W. Wright, in which we have determined that SV40 large T-antigen expression is required for cell division in cells postimmortalization (Wright et al., 1989), and have identified some of the functions o f t antigen that are involved. In addition, the assignment of immortal cell lines to specific groups allowed us to take a focused approach to identify the chromosomes and eventually the genes involved in cellular senescence.
MICROCELL-MEDIATED CHROMOSOME TRANSFER STUDIES We proceeded with the relatively new technique of microcell fusion, introducing normal single human chromosomes into immortal cell lines representing the various complementation groups, to determine an effect on cell division. We began our studies with chromosome 11. The microcell donor was a monochromosomal mouse-human hybrid that retained a single human derivative chromosome 11 in which most of chromosome 11 had translocated to a small portion of the X chromosome, retaining the hypoxanthine guanine phosphoribosyl transferase (HPRT) gene. We introduced this chromosome into H P R T - cell lines representative of 3 of 4 of the complementation groups (Group C cell lines could not be used because H P R T - subclones were not obtained after many attempts). Chromosome 11 had no effect on the proliferation of the ceils, indicating that genes on this chromosome were not involved in cellular senescence (Ning et al., 1991).
522
O.M.PEREIRA-SMITHANDY. NING
However, when we introduced an intact normal human chromosome 4 into cell lines representing the various complementation groups, senescence was induced in HeLa cells (Group B), but not cell lines assigned to the other groups (unpublished results). Work is in progress to determine whether this effect occurs in other cell lines within this group, and to identify the region of the chromosome that is involved in cellular senescence. CONCLUSIONS In summary, we believe that our data clearly demonstrate the dominance of the senescence phenotype and support the hypothesis that cellular aging is an active, genetically based process rather than the result of accumulation of random damage. The microcell fusion results indicate that it will be feasible to use this approach to identify the genes involved in cellular senescence. Once the genes are identified, their regulation and involvement in senescence of various normal cells, such as vascular endothelial cells and T lymphocytes, can be examined. It may then be possible to outline strategies for intervention in the aging individual to, achieve the goal of gerontologists: improvement in the quality of life of the elderly. Acknowledgments - - This work was supported by National Institutes of Health grants AGO5333 and POIAGO7123, and the Pilgeram Fund for Aging Research.
REFERENCES GOLDSTEIN, S. Replicative senescence: The human fibroblast comes of age. Science 249, 1129-1133, 1990. HAYFLICK, L. The limited in vitro lifetime of human diploid cell strains. Exp. Cell Res. 37, 614-636, 1965. HAYFLICK, L. and MOORHEAD, P.S. The serial cultivation of human diploid strains. Exp. CellRes. 25, 585621, 1961. NING, Y., SHAY, J.W., LOVELL, M., TAYLOR, L., LEDBETTER, D.H., and PEREIRA-SMITH, O.M. Tumor suppression by chromosome 11 is not due to cellular senescence. Exp. CellRes. 192, 220-226, 1991. PEREIRA-SMITH, O.M., ROBETORYE, S., NING., Y., and ORSON, F. Hybrids from fusion of normal human T lymphocytes with immortal human cells exhibit limited lifespan. J. Cell. Physiol. 144, 546-549, 1990a. PEREIRA-SMITH, O.M. and SMITH, J.R. Expression of SV40 T antigen in finite lifespan hybrids of normal SV40 transformed cells. Som. Cell Genet. 7, 411-421, 1981. PERERIA-SMITH, O.M. and SMITH, J.R. Evidence for the recessive nature of cellular immortality. Science 221,964-966, 1983. PEREIRA-SMITH, O.M. and SMITH, J.R. Functional simian virus 40 T antigen is expressed in hybrid cells having finite proliferative potential. Mol. Cell. Biol. 7, 1541-1544, 1987. PEREIRA-SMITH, O.M. and SMITH, J,R. Genetic analysis of indefinite division in human cells: Identification of four complementation groups. Proc. Natl. Acad. Sci. U. S. A. 85, 6042-6046, 1988. PEREIRA-SMITH, O.M., STEIN, G.H., ROBETORYE, S., and MEYER-DEMAREST, S. Immortal phenotype of the HeLa variant D98 is recessive in hybrids formed with normal human fibroblasts..L Cell. Physiol. ' 143,222-225, 1990b. WRIGHT, W.E., PEREIRA-SMITH, O.M., and SHAY, J.W. Reversible cellular senescence: Implications for immortalization of normal human diploid fibroblasts. Mol. Cell. Biol. 9, 3088-3092, 1989.