Ask most scientists to describe p21 and their first response would correctly be that it is a cyclin dependent kinase inhibitor induced by the p53 tumour suppressor. Some respondents might add that p21 is also induced independently of p53 during cellular differentiation. The more well informed would remember that one of its earliest described functions was also to bind Proliferating Cell Nuclear Antigen (PCNA), resulting in the direct inhibition of DNA-replication. Many, however, would probably be surprised at the diversity of its interactions and the range of subsequent effects, not just on the cell cycle but also on apoptosis, the stress response, and tumorigenesis. 1 The report by Zhu and co-workers, in this edition of Cell Cycle, provides an opportunity to discuss one of these less familiar, but possibly very important, aspects of p21’s behavior, namely its ability to function as a highly specific regulator of gene expression.

At one level, the suggestion that p21 will regulate transcription is obvious: p21 can inhibit the kinase activity of Cyclin/CDK complexes and these have been shown to regulate the activity of a number of transcription factors. The best example of this straightforward mechanism is found with members of the E2F and Retinoblastoma (Rb) families of transcriptional regulators. Phosphorylation by CDKs relieves repression of E2F complexes by Rb family members during cell cycle progression. Expression of p21 will reverse this effect leading to repression of E2F regulated genes. 1 This is not the whole story, however. Mutants of p21 that no longer inhibit Cyclin/CDK activity are still capable of inhibiting E2F transactivation. 2 Interestingly, this study also reported that p21 could bind directly to E2F-1 and its partner DP-1. Furthermore, when targeted to promoters through fusion to the Gal4 DNA-binding domain, p21 was itself capable of repressing transcription. These latter observations have to be viewed with some caution, since the in vivo significance of the interaction with E2F-1/DP-1 was not proven and Gal4 fusions can be notoriously misleading. Nonetheless, these results demonstrated that p21 is capable of affecting transcription by mechanisms other than merely acting as a Cyclin/CDK inhibitor. So how specific are the transcriptional effects of p21 and can they all be ascribed to regulation of E2F/Rb activity? To address this important question, the Roninson laboratory performed a cDNA microarray analysis using HT1080 human fibrosarcoma cells containing a stably transfected, IPTG inducible, p21 gene. 3 A total of seventy-seven genes were identified as being repressed upon p21 expression. As might be expected, most of these genes are known to be involved in mitosis, DNA-replication, segregation and repair. Many of these genes also contain E2F sites in their promoters suggesting that p21 repressed transcription through the mechanisms discussed above. A number of these genes were not obviously E2F targets, however. Following up on this observation, Zhu et al in this edition of Cell Cycle, have investigated the ability of p21 to inhibit the expression of the genes encoding Polo-like kinase 1 (PLK1) and Topoisomerase IIα (TOPO IIα). Both of the promoters of these genes contain a bipartite motif, designated the CDE-CHR element, that has previously been shown to mediate G1 phase repression of a number of cell cycle regulated genes. 4 In the case of the PLK1 promoter, mutation of either component of this element virtually abolished the ability of p21 to repress transcription. In addition, a CCAAT box just upstream of the CDE-CHR motifs also contributed to this repression. Interestingly, the PLK1 promoter does contain an E2F binding site but …