Replicative aging was first described 40 years ago, when Hayflick (1) reported that normal human lung fibroblasts did not divide indefinitely like cancer cells but rather exhibited a limited ability to proliferate in culture. This paper contradicted the prevailing concept that all cells were immortal, and it was met with great skepticism at the time. Most of the criticism centered on the artificial nature of the in vitro culture environment and the likelihood that inadequate culture conditions were preventing the long-term proliferative capacity of the cells (see Refs. 2 and 3 for recent summaries of these views). As we shall discuss below, this criticism is valid in a variety of culture systems, so that we now must reevaluate some of the data that we had thought reflected replicative aging. However, the identification of telomeric shortening as the mitotic clock that counts cell divisions, the demonstration that telomeres shorten in vivo with donor age, the proof that telomerase can prevent telomeric shortening and immortalize cells, and the observation that most cancer cells have escaped these controls by upregulating/reactivating telomerase have all combined firmly to establish replicative aging as a legitimate in vivo biological process. This chapter will review the development of our current concepts of the mechanisms regulating replicative aging and discuss some of the outstanding issues facing the field.


The two-stage model of cellular senescence, in which Ml (mortality stage 1) and M2 (mortality stage 2) represent independent mechanisms limiting the proliferative capacity of normal cells, was derived to explain the behavior of human cells trans-fected with an inducible simian virus 40 (SV40) large T antigen (4). Various viral oncoproteins (T antigen from SV40, E6/E7 from the high-risk strains of human papilloma viruses, and E1A/E1B from adenovirus type 5) can extend the cultured life span of normal human fibroblasts significantly, but do not directly immortalize the cells. Rather than entering a period of prolonged quiescence, as do normal cells at the limit of their proliferative capacity, cells expressing such proteins enter crisis. An ongoing process of cell division and apoptosis characterizes crisis, so that the population size initially ceases to increase and eventually declines to the point that the culture is frequently lost. Occasionally, as a very rare event, an immortal cell emerges from crisis. An interesting behavior was observed when IMR90 human embryonic lung fibroblasts were immortalized using a dexamethasone-inducible large T antigen. The immortal cells stopped dividing when T antigen was deinduced by the removal of dexamethasone; this suggested that the continued expression of T antigen was necessary for the maintenance of the immortal state. However, the cells also stopped dividing when T antigen was deinduced during the period of extended life span prior to crisis when the cells were not yet immortal.

The model we proposed to explain these observations postulates that senescence initially is caused by the first of two independent mechanisms, mortality stage 1 or the Ml mechanism. Although early studies suggested that the Ml mechanism required the actions of p53 and pRB (4-6), recent observations have raised the possibility that the involvement of the RB pathway reflects a stress response to inadequate culture conditions (see Sec. V), and that p53 is the important cellular sensor of M1. Tumor virus proteins like SV40 T antigen bind to both p53 and pRB and block their activity, thus bypassing M1 and extending the life span of the entire population of cells until an independent second mechanism, M2, is activated. M2 causes crisis, and is not directly affected by the presence of T antigen or similar viral proteins. If a critical M2 gene becomes inactivated, the cells can escape crisis and become immortal. However, these immortal cells have a perfectly functional M1 mechanism, which was never mutated but simply blocked by the presence of T antigen. Consequently, if T antigen is removed any time after the M1 mechanism has been induced, the cells stop dividing due to the actions of M1-activated p53 and pRB.

0 0

Post a comment