CELL CYCLE:
The Expanding Role of Cell Cycle Regulators

The cell cycle clock orchestrates the progression of eukaryotic cells through their growth and division cycles. Much of its importance derives from its job as the master controller of a cell's decision to continue proliferating or to withdraw from the cycle and enter a state of quiescence. But recent work--including that of Di Cunto et al. on page 1069 of this issue (1)--points to a wider role for components of the clock apparatus, some of which extend their reach as far as cellular differentiation.

The core clock machinery is assembled from modular components--cyclins and cyclin-dependent kinases (CDKs). The CDKs control various cellular responses through their ability to phosphorylate appropriate substrates within the cell. The cyclins, acting like guide dogs, bind to and direct CDKs to appropriate substrates during specific phases of the cell cycle, thereby dictating when and where these substrates will become phosphorylated.

The retinoblastoma protein, pRB, and related family members are critical targets for cyclins and CDKs. During the mid-G₁ phase of the growth cycle, their phosphorylation by certain cyclin D-CDK4 and cyclin D-CDK6 complexes enables the activation of yet other cyclin-CDK complexes and the transcription of genes required for S phase entry and progression. All these events permit the cell to advance into the late G₁ and S phases (see the figure).

Equally important components of this core machinery are two groups of CDK inhibitors (CKIs) that block the actions of specific cyclin-CDK complexes (green squares in the figure). In so doing, they may prevent cell cycle progression or induce cells to exit the active proliferative cycle and enter the quiescent G₀ phase. For example, CKIs of the INK4 group (p15, p16, p18, p19) are specialized to block the cyclin-CDK4 and cyclin-CDK6 complexes that are essential to pRB phosphorylation and the associated advance into the late G₁ phase of the cell cycle.

But some of the components of the cell cycle clock have other functions besides direct control of proliferation. The Di Cunto et al. report indicates that the p21 CKI, which is capable of inhibiting a wide spectrum of CDKs operating throughout the cell cycle, also participates in the development of differentiated phenotypes of keratinocytes of the skin. These authors demonstrate that the amount of p21 protein decreases as keratinocytes initiate end-stage differentiation, and that forced expression of p21 can inhibit the differentiation process. Biochemical and mutational evidence indicates that this differentiation-inhibiting function of p21 can be separated from its abilities to inhibit cyclin-CDK complexes (1).

An early precedent for a double life for CKIs has come from the Far1p protein of the budding yeast Saccharomyces cerevisiae. Originally discovered as a CKI induced by mating pheromones, Far1p was later shown to have a distinct function: orienting the yeast cell toward its mating partner (2, 3). Similarly, the mammalian p21 protein studied by Di Cunto et al. has another personality, a domain capable of binding to the PCNA (proliferating cell nuclear antigen) component of DNA polymerases, thereby affecting the process of DNA replication (4, 5). This function of p21 outside of the core clock machinery provides an additional precedent for a multifunctional CKI that can affect cellular targets other than the core components of the clock machinery.

Other surprises of this sort have emerged recently. Cyclin D1 was initially portrayed as an important activator of the CDK4 and CDK6 complexes that phosphorylate pRB and related proteins in the G₁ phase (6). But reports from two groups indicate, totally unexpectedly, that cyclin D1 can bind and activate the estrogen receptor (ER) (7, 8). Before this work, estrogen was thought to be the major physiologic activator of this receptor. The biological consequences of the cyclin D1-ER interaction remain unclear; given the wide-ranging actions of the ER, some of them might involve differentiation-like responses.

pRB has been portrayed exclusively as the brake shoe of cell cycle advance in the G₁ phase of the growth cycle; its absence or functional inactivation in many types of human tumors is compatible with this action (9). But new research indicates that pRb helps to direct the development of at least two distinct differentiation programs. Cultured myoblasts do not differentiate properly in the absence of pRB (10, 11). This differentiation function appears to be associated with a domain of pRB that is distinct from those domains that directly control proliferation (12). Yet other work indicates an analogous role for pRB in programming adipocyte differentiation (13). Although these results stem from in vitro differentiation models, we suspect that they reflect processes operative in living tissues and that the differentiation programs in a variety of other tissues may be similarly dependent on pRB function.

A particularly intriguing example of an intrinsic cell cycle regulator moonlighting in another cellular function is the CDK-activating enzyme CAK, a kinase required for the full stimulation of CDK activity. In mammalian cells, CAK is also a critical component of the RNA polymerase holoenzyme (its TFIIH subunit), required for the transcription of most cellular genes (14-16). Whether this is an example of a cell cycle regulator being co-opted by evolution to perform a transcriptional function or the reverse is not known.

The portrait of the cell cycle clock as an apparatus focused exclusively on governing proliferation has become simplistic. It now seems clear that this apparatus, embedded in the heart of the eukaryotic cell for a billion years, has been exploited by the tinkering hand of evolution to control other important cellular functions, particularly those that are required for complex cellular differentiation. Evolution, always opportunistic, uses the hardware already lying around on the shelves to make clever new toys. The powers of the cell cycle clock apparatus are likely to be far broader than currently suspected.

References

1. F. Di Cunto et al., Science 280, 1069 (1998).
2. R. Dorer et al., J. Cell Biol. 131, 845 (1995) [Medline].
3. N. Valtz, M. Peter, I. Herskowitz, ibid., p. 863 [Medline].
4. S. Waga et al., Nature 369, 574 (1994) [Medline].
5. H. Flores-Rozas et al., Proc. Natl. Acad. Sci. U.S.A. 91, 8655 (1994) [Medline].
6. For a review, see C. J. Sherr, Cell 79, 551 (1994) [Medline].
7. R. M. Zwijsen et al., ibid. 88, 405 (1997) [Medline].
8. E. Neuman et al., Mol. Cell. Biol. 17, 5338 (1997) [Medline].
9. For a review, see R. A. Weinberg, Science 254, 1138 (1991) [Medline].
10. W. Gu et al., Cell 72, 309 (1993) [Medline].
11. B. G. Novitch, G. J. Mulligan, T. Jacks, A. B. Lassar, J. Cell Biol. 135, 441 (1996) [Medline].
12. W. R. Sellers et al., Genes Dev. 12, 95 (1998) [Medline].
13. P. L. Chen, D. J. Riley, Y. Chen, W. H. Lee, ibid. 10, 2794 (1996) [Medline].
14. W. J. Feaver, J. Q. Svejstrup, N. L. Henry, R. D. Kornberg, Cell 79, 1103 (1994) [Medline].
15. H. Serizawa et al., Nature 374, 280 (1995) [Medline].
16. R. Shiekhattar et al., ibid., p. 283 [Medline].