An analysis of the regulation of DNA synthesis by cdk2, Cip1, and licensing factor

Two divergent types of nerve pathology in patients with different P0 mutations in Charcot-Marie-Tooth disease

In seven unrelated patients with a demyelinating motor and sensory neuropathy, we found mutations in exons 2 and 3 of the P0 gene. Morphologic examination of sural nerve biopsy specimens showed a demyelinating process with onion bulb formation in all cases. In four patients, ultrastructural examination demonstrated uncompacted myelin in 23 to 68% of the myelinated fibers, which is in agreement with the widely accepted function of P0 as a homophilic adhesion molecule. Three patients showed normal compact myelin, but morphology was dominated by the abundant occurrence of focally folded myelin. The two divergent pathologic phenotypes exemplify that some mutations act differently on P0 protein formation or function than others, which is probably determined by site and nature of the mutation in the P0 gene.

Sequence and context effects on origin function in mammalian cells

Jacob and Brenner proposed a model for control of DNA replication in which a trans-acting initiator protein binds to a cis-acting replicator to effect initiation of nascent DNA chains at a fixed locus. Although replicators have been identified in prokaryotic and simple eukaryotic genomes, it has been much more difficult to demonstrate their presence in mammalian chromosomes. Owing to the lack of genetic approaches for identifying mammalian replicators, investigators have directed attention to localizing nascent strand start sites, which should lie close to replicators. Toward this end, a variety of clever techniques have been invented for analyzing replication intermediates, but only rarely have more than one of these techniques been applied to a single locus. However, virtually all have been used to analyze the dihydrofolate reductase locus in CHO cells. The picture that has developed in this locus is that initiation can occur at any of a large number of sites scattered throughout a broad zone, but somewhat more frequently near two sites that may correspond to true genetic replicators. Furthermore, it appears that local transcriptional activity, as well as appropriate torsional stress (as imparted by local attachment to the nuclear matrix), may have profound effects on origin activity.

Organization of DNA into foci during replication

In metazoan cells, DNA replication during the S phase of the cell cycle takes place at discrete locations within the nucleus. These sites, or foci, appear to participate in clustering many replicons together and synchronously regulating the activation of these replicon units. Consistent with this role. many replication proteins have been observed to attach to foci during the S phase of the cell cycle. Recently, cell-free replication extracts have been used both to characterize the events that are involved in either the formation of these sites or in the regulation of foci assembly, and to purify candidate proteins which may be integral core structural proteins responsible for forming foci. These advances provide a foundation for investigating foci structure and function at a biochemical level.

Nuclear proteins of quiescent Xenopus laevis cells inhibit DNA replication in intact and permeabilized nuclei

Quiescent cells from adult vertebrate liver and contact-inhibited or serum-deprived tissue cultures are active metabolically but do not carry out nuclear DNA replication and cell division. Replication of intact nuclei isolated from either quiescent Xenopus liver or cultured Xenopus A6 cells in quiescence was barely detectable in interphase extracts of Xenopus laevis eggs, although Xenopus sperm chromatin was replicated with approximately 100% efficiency in the same extracts. Permeabilization of nuclei from quiescent Xenopus liver or cultured Xenopus epithelial A6 cells did not facilitate efficient replication in egg extracts. Moreover, replication of Xenopus sperm chromatin in egg extracts was strongly inhibited by a soluble extract of isolated Xenopus liver nuclei; in contrast, complementary-strand synthesis on single-stranded DNA templates in egg extracts was not affected. Inhibition was specific to endogenous molecules localized preferentially in quiescent as opposed to proliferating cell nuclei, and was not due to suppression of cdk2 kinase activity. Extracts of Xenopus liver nuclei also inhibited growth of sperm nuclei formed in egg extracts. However, the rate and extent of decondensation of sperm chromatin in egg extracts were not affected. The formation of prereplication centers detected by anti-RP-A antibody was not affected by extracts of liver nuclei, but formation of active replication foci was blocked by the same extracts. Inhibition of DNA replication was alleviated when liver nuclear extracts were added to metaphase egg extracts before or immediately after Ca++ ion-induced transition to interphase. A plausible interpretation of our data is that endogenous inhibitors of DNA replication play an important role in establishing and maintaining a quiescent state in Xenopus cells, both in vivo and in cultured cells, perhaps by negatively regulating positive modulators of the replication machinery.

Activation of the Xenopus cyclin degradation machinery by full-length cyclin A

The entry into mitosis is dependent on the activation of mitotic forms of cdc2 kinase. In many cell types, cyclin A-associated kinase activity peaks just prior to that of cyclin B, although the precise role of cyclin A-associated kinase in the entry into mitosis is still unclear. Previous work has suggested that while cyclin B is capable of triggering cyclin destruction in Xenopus cell-free systems, cyclin A-associated kinase is not able to support this function. Here we have expressed a full-length human cyclin A in Escherichia coli and purified the protein to homogeneity by virtue of an N-terminal histidine tag. We have found that when added to Xenopus cell-free extracts free of cyclin B and incapable of protein synthesis, the temporal pattern of cyclin A-associated cdc2 kinase activity showed distinct differences that were dependent on the concentration of cyclin A added. When cyclin A was added to a concentration that generated levels of cdc2 kinase activity capable of inducing nuclear envelope breakdown, the histone H1 kinase activity profile was bi-phasic, consisting of an activation phase followed by an inactivation phase. Inactivation was found to be due to cyclin destruction, which was prevented by mos protein. Cyclin destruction was followed by nuclear reassembly and an additional round of DNA replication, indicating that there is no protein synthesis requirement for DNA replication in this embryonic system. It has been suggested that the evolutionary recruitment of cyclin A into an S phase function may have necessitated the loss of an original mitotic ability to activate the cyclin destruction pathway. The results presented here indicate that cyclin A has not lost the ability to activate its own destruction and that cyclin A-mediated activation of the cyclin destruction pathway permitted destruction of cyclin B1 as well as cyclin A, indicating that there are not distinct cyclin A and cyclin B destruction pathways. Thus the ordered progression of the cell cycle requires the careful titration of cyclin. A concentration in order to avoid activation of the cyclin destruction pathway before sufficient active cyclin B/cdc2 kinase has accumulated.

Chromatin binding, nuclear localization and phosphorylation of Xenopus cdc21 are cell-cycle dependent and associated with the control of initiation of DNA replication

A Xenopus homologue of Schizosaccharomyces pombe cdc21 has been characterized as a new member of the MCM family of proteins. The cdc21 protein exhibits cell-cycle dependent chromatin binding and phosphorylation in association with S-phase control. Cdc21 binds to decondensing chromatin at the end of mitosis, localizing to numerous foci which form prior to reconstitution of the nuclear membrane. The association of cdc21 with chromatin occurs in membrane-free high speed extracts and is resistant to detergent extraction. The spatial organization of the cdc21 foci resembles that of pre-replication centres though no co-localization with RP-A was observed. Cdc21 remains bound to chromatin during the initiation of DNA replication and is displaced as the DNA replication forks progress. These subnuclear changes in localization correlate with cell-cycle-regulated changes in phosphorylation. Cdc21 binds to chromatin in an underphosphorylated state, but in early S phase the nuclear localized cdc21 is partially phosphorylated before it is displaced from the chromatin. Cytoplasmic cdc21 remains underphosphorylated but at the beginning of mitosis the entire pool of cdc21 is hyperphosphorylated, possibly by the cdc2/cyclin B kinase. These properties identify Xenopus cdc21 as a possible component of the DNA licensing factor.

Cell cycle control by Xenopus p28Kix1, a developmentally regulated inhibitor of cyclin-dependent kinases

We have isolated Xenopus p28Kix1, a member of the p21CIP1/p27KIP1/p57KIP2 family of cyclin-dependent kinase (Cdk) inhibitors. Members of this family negatively regulate cell cycle progression in mammalian cells by inhibiting the activities of Cdks. p28 shows significant sequence homology with p21, p27, and p57 in its N-terminal region, where the Cdk inhibition domain is known to reside. In contrast, the C-terminal domain of p28 is distinct from that of p21, p27, and p57. In co-immunoprecipitation experiments, p28 was found to be associated with Cdk2, cyclin E, and cyclin A, but not the Cdc2/cyclin B complex in Xenopus egg extracts. Xenopus p28 associates with the proliferating cell nuclear antigen, but with a substantially lower affinity than human p21. In kinase assays with recombinant Cdks, p28 inhibits pre-activated Cdk2/cyclin E and Cdk2/cyclin A, but not Cdc2/cyclin B. However, at high concentrations, p28 does prevent the activation of Cdc2/cyclin B by the Cdk-activating kinase. Consistent with the role of p28 as a Cdk inhibitor, recombinant p28 elicits an inhibition of both DNA replication and mitosis upon addition to egg extracts, indicating that it can regulate multiple cell cycle transitions. The level of p28 protein shows a dramatic developmental profile: it is low in Xenopus oocytes, eggs, and embryos up to stage 11, but increases approximately 100-fold between stages 12 and 13, and remains high thereafter. The induction of p28 expression temporally coincides with late gastrulation. Thus, although p28 may play only a limited role during the early embryonic cleavages, it may function later in development to establish a somatic type of cell cycle. Taken together, our results indicate that Xenopus p28 is a new member of the p21/p27/p57 class of Cdk inhibitors, and that it may play a role in developmental processes.

Hepatocyte growth factor releases epithelial and endothelial cells from growth arrest induced by transforming growth factor-beta1

Human lung fibroblasts and Mv1Lu mink lung epithelial cells were used as a model to study the role of extracellular matrix in epithelial-mesenchymal interactions. Extracellular matrices of fibroblasts were found to contain growth promoting activity that reduced the sensitivity of Mv1Lu cells to the growth inhibitory effects of transforming growth factor-beta (TGF-beta). The majority of the activity was identified as hepatocyte growth factor/scatter factor (HGF) by inhibition with specific antibodies and by reconstitution of the effect by recombinant HGF. HGF induced cell proliferation when contact-inhibited Mv1Lu cells were trypsinized and plated in the presence of TGF-beta1. The effect was valid also in assays where Madin-Darby canine kidney epithelial cells or bovine capillary endothelial cells were used. The multiplication of chronically TGF-beta1 inhibited Mv1Lu cells was also induced by HGF. In addition, HGF induced anchorage independent growth of Mv1Lu cells that was refractory to TGF-beta1 growth inhibition. Immunoprecipitation analysis indicated that HGF prevented the suppression of Cdk4 and Cdk2, but not the induction of p21, by TGF-beta1. Since both TGF-beta1 and HGF require proteolysis for activation, the results imply that proteolytic activity of epithelial and endothelial cells directs their responses to signals from mesenchymal-type extracellular matrices, and that during development, matrix-bound growth and invasion promoting and suppressing factors are activated in a coordinated manner.

The effect of 6-dimethylaminopurine (6-DMAP) on DNA synthesis in activated mammalian oocytes

Mouse, sheep and bovine metaphase II oocytes were briefly preincubated in medium with 6-dimethylaminopurine (6-DMAP) and thereafter activated with ethanol before further culture in 6-DMAP supplemented medium. The presence of 6-DMAP enhances the efficiency of activation and the speed of pronuclear formation but has no effect on DNA synthesis. These results are discussed in the context of recently published data showing that the preincubation of condensed chromosomes with 6-DMAP blocks DNA synthesis when nuclei become reformed.

Role for a Xenopus Orc2-related protein in controlling DNA replication

The six-subunit origin recognition complex (ORC) is essential for the initiation of DNA replication at specific origins in the budding yeast Saccharomyces cerevisiae. An important issue is whether DNA replication in higher eukaryotes, in which the characteristics of replication origins are poorly defined, occurs by an ORC-dependent mechanism. We have identified a Xenopus laevis Orc2-related protein (XORC2) by its ability to rescue a mitotic-catastrophe mutant of the fission yeast Schizosaccharomyces pombe. We show that immunodepletion of XORC2 from Xenopus egg extracts abolishes the replication of chromosomal DNA but not elongation synthesis on a single-stranded DNA template. Indirect immunofluorescence indicates that XORC2 binds to chromatin well before the commencement of DNA synthesis, and even under conditions that prevent the association of replication licensing factor(s) with the DNA. These findings suggest that Orc2 plays an important role at an early step of chromosomal replication in animal cells.

Cdk2 kinase is required for entry into mitosis as a positive regulator of Cdc2-cyclin B kinase activity

In higher eukaryotes, Cdk2 kinase plays an essential role in regulating the G1-S transition. Here, we use cycling Xenopus egg extracts to examine the requirement of Cdk2 kinase on progression into mitosis. Interestingly, when Cdk2 kinase activity is inhibited by the Cdk-specific inhibitor, p21Cip1, a block to mitosis occurs, and inactive Cdc2-cyclin B accumulates. This block occurs in the absence of nuclei and is not due to direct inhibition of Cdc2 by Cip. Importantly, this block to mitosis is reversible by restoring Cdk2-cyclin E kinase activity to a Cip-treated cycling extract. Moreover, immunodepletion of Cdk2 from interphase extracts prevents activation of Cdc2 upon the addition of exogenous cyclin B. Thus, our data show that Cdk2 kinase is a positive regulator of Cdc2-cyclin B complexes and establish a link between Cdk2 kinase and cell cycle progression into mitosis.

FFA-1, a protein that promotes the formation of replication centers within nuclei

In the nuclei of eukaryotic cells, initiation of DNA replication occurs at a discrete number of foci. One component of these foci is the DNA replication factor RP-A. Here, the process leading to the association of RP-A with foci was reconstituted with cytosolic fractions derived from Xenopus eggs. With the use of this fractionated system, a 170-kilodalton protein required for the assembly of RP-A into foci was identified and purified. The protein appears to be an integral component of the foci at which replication of DNA is initiated in eukaryotic nuclei.

DNA replication. Almost licensed

Exact duplication of all the DNA in a cell occurs during each S phase, and only once in each cell cycle. Recent results show that conserved proteins of the MCM family contribute to these precisely regulated events.

RNA polymerase III transcription in synthetic nuclei assembled in vitro from defined DNA templates

Although much is known of the basic control of transcription, little is understood of the way in which the structural organization of the nucleus affects transcription. Synthetic nuclei, assembled de novo in extracts of Xenopus eggs, would be predicted to have a large potential for approaching the role of nuclear structure in RNA biogenesis. Synthetic nuclei provide a system in which the genetic content of the nuclei, as well as the structural and enzymatic proteins within the nuclei, can be manipulated. In this study, we have begun to examine transcription in such nuclei by using the most simple of templates, RNA polymerase III (pol III)-transcribed genes. DNA encoding tRNA or 5S genes was added to an assembly extract, and nuclei were formed entirely from the pol III templates. Conditions which allowed nuclear assembly and pol III transcription to take place efficiently and simultaneously in the assembly extract were found. To examine whether pol III transcription could initiate within synthetic nuclei, or instead was inhibited in nuclei and initiated only on rare unincorporated templates, we identified transcriptional inhibitors that were excluded from nuclei. We found that these inhibitors, heparin and dextran sulfate, blocked pol III transcription in the absence of assembly but did not do so following nuclear assembly. At the concentrations used, the inhibitors had no deleterious effect on nuclear structure itself or on nuclear import. We conclude that pol III transcription is active in synthetic nuclei, and this conclusion is further strengthened by the finding that pol III transcripts could be coisolated with synthetic nuclei. The rapid and direct transcriptional analysis possible with pol III templates, coupled with the simple experimental criteria developed in this study for distinguishing between nuclear and non-nuclear transcription, should now allow a molecular analysis of the effect of nuclear structure on transcriptional and posttranscriptional control.

Early events in DNA replication require cyclin E and are blocked by p21CIP1

Using immunodepletion of cyclin E and the inhibitor protein p21WAF/CIP1, we demonstrate that the cyclin E protein, in association with Cdk2, is required for the elongation phase of replication on single-stranded substrates. Although cyclin E/Cdk2 is likely to be the major target by which p21 inhibits the initiation of sperm DNA replication, p21 can inhibit single-stranded replication through a mechanism dependent on PCNA. While the cyclin E/Cdk2 complex appears to have a role in the initiation of DNA replication, another Cdk kinase, possibly cyclin A/Cdk, may be involved in a later step controlling the switch from initiation to elongation. The provision of a large maternal pool of cyclin E protein shows that regulators of replication are constitutively present, which explains the lack of a protein synthesis requirement for replication in the early embryonic cell cycle.