CDK inhibitors are thought to prevent cell proliferation by negatively regulating cyclin-CDK complexes. We propose that the opposite is also true, that cyclin-CDK complexes in mammmalian cells can promote cell cycle progression by directly down-regulating CDK inhibitors. We show that expression of cyclin E-CDK2 in murine fibroblasts causes phosphorylation of the CDK inhibitor p27 Kip1 on T187, and that cyclin E-CDK2 can directly phosphorylate p27 T187 in vitro. We further show that cyclin E-CDK2-dependent phosphorylation of p27 results in elimination of p27 from the cell, allowing cells to transit from G1 to S phase. Moreover, mutation of T187 in p27 to alanine creates a p27 protein that causes a G1 block resistant to cyclin E and whose level of expression is not modulated by cyclin E. A kinetic analysis of the interaction between p27 and cyclin E-CDK2 explains how p27 can be regulated by the same enzyme it targets for inhibition. We show that p27 interacts with cyclin E-CDK2 in at least two distinct ways: one resulting in p27 phosphorylation and release, the other in tight binding and cyclin E-CDK2 inhibition. The binding of ATP to the CDK governs which state predominates. At low ATP (< 50 ~M) p27 is primarily a CDK inhibitor, but at ATP concentrations approaching physiological levels (> 1 mM) p27 is more likely to be a substrate. Thus, we have identified p27 as a biologically relevant cyclin E-CDK2 substrate, demonstrated the physiological consequences of p27 phosphorylation, and developed a kinetic model to explain how p27 can be both an inhibitor and a substrate of cyclin E-CDK2.

Cyclin E is a mammalian G 1 cyclin that is both required and rate limiting for entry into S phase. The expression of cyclin E is periodic, peaking at the GrS transition and then decaying as S phase progresses. To understand the mechanisms underlying cyclin E periodicity, we have investigated the regulation of cyclin E degradation. We find that cyclin E is degraded by the ubiquitin-proteasome system, and that this degradation is regulated by both cdk2 binding and cdk2 catalytic activity. Free cyclin E is readily ubiquitinated and degraded by the proteasome. Binding to cdk2 protects cyclin E from ubiquitination, and this protection is reversed by cdk2 activity in a process that involves phosphorylation of cyclin E itself. The data are most consistent with a model in which cdk2 activity initiates cyclin E degradation by promoting the disassembly of cyclin E-cdk2 complexes, followed by the ubiquitination and degradation of free cyclin E.[Key Words: Cyclin; cdk; ubiquitin; proteasome] Received June 7, 1996; accepted July 8, 1996.Cyclins, in association with their catalytic subunits, the cyclin-dependent kinases (cdks), drive the eukaryotic cell cycle through key transitions by phosphorylating a group of largely unknown substrates. More than 10 mammalian cyclins have been identified to date, and most (but not all) are expressed periodically. Each phase of the cell cycle has a unique profile of cyclin-cdk activity, and these distinct fluctuations in cyclin activity are essential for normal cell-cycle progression. The molecular mechanisms underlying cyclin periodicity involve both transcriptional and post-transcriptional regulation. Many cyclin mRNAs exhibit cell-cycle dependent fluctuations, and in some cases, specific transcription factors have been implicated in controlling this periodicity. For example, cyclin E is controlled by E2F (DeGregori et al.

The Cdk inhibitor p21Cip1 is an unstable protein. Pharmacologic inhibition of the proteasome increases the half-life of p21 from less than 30 min to more than 2 hr and results in the accumulation of p21-ubiquitin conjugates. To determine whether ubiquitination was required for proteasomal degradation of p21, we constructed mutant versions of p21 that were not ubiquitinated in vivo. Remarkably, these mutants remained unstable and increased in abundance upon proteasome inhibition, indicating that direct ubiquitination of p21 is not necessary for its turnover by the proteasome. The frequently observed correlation between protein ubiquitination and proteasomal degradation is insufficient to conclude that ubiquitination is a prerequisite for degradation.

The mechanism by which calf thymus DNA primase synthesizes RNA primers was examined. Primase first binds a single-stranded DNA template (KD << 100 nM) and can then slide along the DNA in order to find a start for initiating primer synthesis. NTP binding appears ordered, such that the NTP which eventually becomes the second nucleotide of the primer binds the E.DNA complex first. The NTP that becomes the second nucleotide of the primer thereby influences where primase initiates. Primer synthesis is remarkably slow (0.0027 s-1 at 20 microM NTP). The rate-limiting step is after formation of the E.DNA.NTP.NTP complex and before or during dinucleotide synthesis. After synthesis of the dinucleotide, additional NTPs are rapidly polymerized. Primase products are 2-10 nucleotides long. If the enzyme fails to synthesize a primer at least 7 nucleotides long, it reinitiates rather than dissociating from the template. Once a primer at least 7 nucleotides long has been generated, however, subsequent primase activity is inhibited. This inhibition is due to the generation of a stable primer-template complex, which likely remains associated with pol alpha.primase. The role of primase is to synthesize primers that pol alpha can elongate. The ability of primase to distinguish between primers at least 7 nucleotides long and shorter products therefore likely reflects the fact that pol alpha only utilizes primers at least 7 nucleotides long.