Faculty Bibliography

Cyclin E was first identified by screening human cDNA libraries for genes that would complement G1 cyclin mutations in Saccharomyces cerevisiae and has subsequently been found to have specific biochemical and physiological properties that are consistent with it performing a G1 function in mammalian cells. Most significantly, the cyclin E-Cdk2 complex is maximally active at the G1/S transition, and overexpression of cyclin E decreases the time it takes the cell to complete G1 and enter S phase. We have now found that mammalian cells express two forms of cyclin E protein which differ from each other by the presence or absence of a 15-amino-acid amino-terminal domain. These proteins are encoded by alternatively spliced mRNAs and are localized to the nucleus during late G1 and early S phase. Fibroblasts engineered to constitutively overexpress either form of cyclin E showed elevated cyclin E-dependent kinase activity and a shortened G1 phase of the cell cycle. The overexpressed cyclin E protein was detected in the nucleus during all cell cycle phases, including G0. Although the cyclin E protein could be overexpressed in quiescent cells, the cyclin E-Cdk2 complex was inactive. It was not activated until 6 to 8 h after readdition of serum, 4 h earlier than the endogenous cyclin E-Cdk2. This premature activation of cyclin E-Cdk2 was consistent with the extent of G1 shortening caused by cyclin E overexpression. Microinjection of affinity-purified anti-cyclin E antibodies during G1 inhibited entry into S phase, whereas microinjection performed near the G1/S transition was ineffective. These results demonstrate that cyclin E is necessary for entry into S phase. Moreover, we found that cyclin E, in contrast to cyclin D1, was required for the G1/S transition even in cells lacking retinoblastoma protein function. Therefore, cyclins E and D1 control two different transitions within the human cell cycle

The E6 protein of the high-risk human papillomaviruses inactivates the tumor suppressor protein p53 by stimulating its ubiquitinylation and subsequent degradation. Ubiquitinylation is a multistep process involving a ubiquitin-activating enzyme, one of many distinct ubiquitin-conjugating enzymes, and in certain cases, a ubiquitin ligase. In human papillomavirus-infected cells, E6 and the E6-associated protein are thought to act as a ubiquitin-protein ligase in the ubiquitinylation of p53. Here we describe the cloning of a human ubiquitin-conjugating enzyme that specifically ubiquitinylates E6-associated protein. Furthermore, we define the biochemical pathway of p53 ubiquitinylation and demonstrate that in vivo inhibition of various components in the pathway leads to an inhibition of E6-stimulated p53 degradation.

p16Ink4 (inhibitor of cyclin-dependent kinase 4) is a cell cycle regulator that specifically binds to and inhibits Cdk4. Recently, the human mts1 (multiple tumor suppressor 1) gene, deleted or mutated in various primary tumors and in a large number of transformed cell lines, was found to be identical to ink4. In this study we have surveyed by immunoblotting the protein levels of p16Ink4 in normal and transformed human cells. We determined that p16Ink4 was differentially expressed in diploid cells derived from different tissues, in contrast to another cell cycle inhibitor, p21Waf1, which is ubiquitously expressed. In some tumor cell lines p16Ink4 protein was not detected, presumably because of a homozygous deletion of its gene. By contrast, it was found to be overexpressed in other cell lines when compared to levels in their normal counterparts. Interestingly, high levels of p16Ink4 protein correlated with functional inactivation of the retinoblastoma gene product. We also found that p16Ink4 protein expression varies during the cell cycle peaking during S phase. These results show a functional relationship between p16Ink4 and the retinoblastoma gene product and indicate that p16Ink4 is required for Cdk4 inhibition only at the G1-S transition at the time when Cdk4 kinase activity is no longer necessary

In this study we have surveyed by immunoblotting the protein levels of Cyclin D1, D2, D3 and their catalytic partners, Cdk4 and Cdk6 in normal and transformed human cells. We found that all these proteins were differentially expressed in diploid cells derived from different tissues, in contrast to Cyclin E, Cyclin A and Cdk2 which are ubiquitously expressed. D-type Cyclins were never dramatically overexpressed and often very poorly expressed in tumor cell lines when compared to the levels in their normal counterparts. In contrast, Cdk4 was expressed at high levels in several tumor cell lines and Cdk6 was ectopically expressed in two sarcoma lines, suggesting a possible involvement of these two Cdks in oncogenesis. Interestingly, low levels of Cyclin D1 and D3 proteins always correlated with functional inactivation of the retinoblastoma gene product (pRb). In cells displaying active pRb, Cyclin D1 was found associated with Cdk4 regardless of whether the p53 gene was wild-type or mutant. Microinjection during G1 of Cyclin D1 anti-sense cDNA or anti-Cyclin D1 antibody in these cells arrested the cell cycle in G1. In cells lacking pRb function, Cyclin D1 was dissociated from Cdk4. Microinjection during G1 of Cyclin D1 antisense cDNA or anti-Cyclin D1 antibody in these cells did not affect G1 progression. These results show that (i) in the absence of pRb, Cyclin D1 is expressed at low levels, is dissociated from Cdk4 and becomes dispensable in G1; (ii) Cyclin D1 needs to be associated with its catalytic subunit, Cdk4, to function as a positive regulator of G1 progression

Cyclin D1 is a key regulator of the G1 phase of the cell cycle. Inhibition of cyclin D1 function results in cell cycle arrest, whereas unregulated expression of the protein accelerates G1. Cyclin D1 is localized to the nucleus during G1. We found that during repair DNA synthesis, subsequent to UV-induced DNA damage, G1 cells readily lost their cyclin D1 while the proliferating cell nuclear antigen (PCNA) tightly associated with nuclear structures. Microinjection of cyclin D1 antisense accelerated DNA repair, whereas overexpression of cyclin D1 prevented DNA repair and the relocation of PCNA after DNA damage. Coexpression of cyclin D1 with its primary catalytic subunit, Cdk4, or with Cdk2, also prevented repair. In contrast, coexpression of PCNA, which is also a cyclin D1-associated protein, restored the ability of cells to repair their DNA. Acute overexpression of cyclin D1 in fibroblasts prevented them from entering S phase. Again, these effects were abolished by coexpression of cyclin D1 together with PCNA, but not with Cdk4 or Cdk2. Altogether, these results indicate that down-regulation of cyclin D1 is necessary for PCNA relocation and repair DNA synthesis as well as for the start of DNA replication. Cyclin D1 appears to be an essential component of a G1-checkpoint

Among the key cell cycle regulators, cyclin D1 has been implicated most strongly in oncogenesis. This G1 cyclin is a putative proto-oncogene whose clonal rearrangement and/or amplification and mRNA overexpression occurs in several types of human neoplasias. We have now raised a series of monoclonal antibodies to human cyclin D1 and analysed its regulation at the protein level in 40 human tumour cell lines. We found that 12 cell lines displayed low or undetectable cyclin D1 protein level, while the remaining lines accumulated the protein to a level comparable to, or moderately higher than, that of four normal diploid non-immortalized cell types. The cell cycle-dependent oscillation and subcellular localization of cyclin D1 were similar in both tumour and normal cells. The protein localized to the nucleus of G1 cells, and it was reduced to immunocytochemically undetectable level in DNA-replicating cells. At the functional level, microinjection and electroporation of anti-D1 antibodies revealed that in most tumour cell lines studied, including those with amplification at the cyclin D1 locus, this cyclin is essential for cell cycle progression in G1. Some tumours, however, seem to have evolved mechanism(s) that enable them to bypass the requirement for functional cyclin D1.

A cascade of events is triggered upon the addition of growth factor to quiescent mammalian cells, which ultimately restarts proliferation by inducing the transition from G0/G1 to S-phase. We have studied cyclin D1, a putative G1 cyclin, in normal diploid human fibroblasts. Cyclin D1 accumulated and reached a maximum level before S-phase upon the addition of serum to quiescent cells. The protein was localized to the nucleus, and it disappeared from the nucleus as cells proceeded into S-phase. Microinjection of anti-cyclin D1 antibodies or antisense plasmid prevented cells from entering S-phase, and the kinetics of inhibition showed that cyclin D1 is required at a point in the cell cycle earlier than cyclin A. These results demonstrate that cyclin D1 is a critical target of proliferative signals in G1.

We investigated the temporal regulation of cyclin A- and B1-dependent kinases in human lymphoma cells treated with nitrogen mustard (HN2) and pentoxifylline, to determine whether the activity of these complexes correlated with cell cycle arrest induced by DNA damage. Cells were synchronized in G1/S, treated with HN2, and then postincubated with pentoxifylline. HN2-induced a protracted delay in G2 phase. This delay correlated with suppression of cyclin B1- and cdc2-kinase activities, and stabilization of hyperphosphorylated-cdc2 in the presence of similar cyclin B1 levels to those found in mitosis. HN2 had no discernible effect on the S phase activity of cyclin A- or cdk2-immune complexes. Entry of control cells into mitosis correlated with destruction of cyclin A, disappearance of cyclin A-bound cdk2 and decreased cdk2 kinase activity. G2 delay induced by HN2 was associated with stabilization of cyclin A, increased abundance of cyclin A-bound cdk2, and increased cdk2 activity. Cyclin A was also associated with cdc2, which, contrary to complexes containing cdk2, were only activated upon entry into mitosis. Pentoxifylline abrogated cell cycle arrest induced by aphidicolin and HN2 in human lymphoma cells. Pentoxifylline also reverted the activity of cyclin A- and B1-kinases in HN2-treated cells to approximately that observed in controls. Our findings suggest that delayed entry into mitosis following DNA damage correlates with suppression of cyclin B1/cdc2 and cyclin A/cdc2 complexes, while maintaining cyclin A/cdc2 complexes in an active state. Furthermore, we found that pentoxifylline disrupts the signal transduction pathway that regulates these complexes when damaged DNA is present, resulting in abrogation of cell cycle arrest.

We have investigated the effects of deregulated expression of the human c-MYC protooncogene on cyclin gene expression and on the transcription factor E2F. We found that constitutive expression of MYC or activation of conditional MycER chimeras led to higher levels of cyclin A and cyclin E mRNA. Activation of cyclin A expression by MYC led to a growth factor-independent association of cyclin A and cdk2 with the transcription factor E2F and correlated with an increase in E2F transcriptional activity. In contrast, expression of the G1 phase cyclin D1 was strongly reduced in MYC-transformed cells. In synchronized cells, repression of cyclin D1 by MYC occurred very early in the G1 phase of the cell cycle.

In mammalian cells inhibition of the cdc2 function results in arrest in the G2-phase of the cell cycle. Several cdc2-related gene products have been identified recently and it has been hypothesized that they control earlier cell cycle events. Here we have studied the relationship between activation of one of these cdc2 homologs, the cdk2 protein kinase, and the progression through the cell cycle in cultured human fibroblasts. We found that cdk2 was activated and specifically localized to the nucleus during S phase and G2. Microinjection of affinity-purified anti-cdk2 antibodies but not of affinity-purified anti-cdc2 antibodies, during G1, inhibited entry into S phase. The specificity of these effects was demonstrated by the fact that a plasmid-driven cdk2 overexpression counteracted the inhibition. These results demonstrate that the cdk2 protein kinase is involved in the activation of DNA synthesis