Melvin L. DePamphilis, PhD, Head, Section on Eukaryotic Gene Regulation
Kotaro Kaneko, PhD, Staff Scientist
Alex Vassilev, PhD, Staff Scientist
Kohji Noguchi, PhD, Visiting Scientist
Matthew Kohn, PhD, Postdoctoral Fellow
Soma Ghosh, PhD, Visiting Fellow
Zakir Ullah, PhD, Visiting Fellow
Rieko Yagi, PhD, Visiting Fellow
Ilian Radichev, BS, Graduate Student
Xiaohong Zhang, BA, Technical Assistant

A fertilized human egg contains 7 pg (about two meters) of DNA, which encode the roughly 27,000 required human genes. After 5 trillion cell divisions, an adult human contains from 10 to 100 trillion cells and over 20 trillion meters (more than 9,300 million miles) of DNA, equivalent to about 100 times the distance from the earth to the sun. Our goal is to understand how such a feat is accomplished and the consequences of errors. When errors occur during DNA replication, the cell either stops further replication while it tries to correct the problem or, if the problem cannot be corrected, the affected cell is programmed to commit suicide, a process called apoptosis. If an error goes undetected by the cell's checkpoint controls, diseases can result from aberrant DNA replication, altering the DNA sequence of a particular gene or its regulatory components, or mutations within a DNA replication or DNA repair gene (at least 80 known). Moreover, at least 40 human diseases are caused by viruses with DNA genomes and many more by microorganisms with DNA genomes. The transmission of these viruses and microorganisms, and in many cases their pathogenesis, depends on their ability to replicate their genome. Therefore, understanding the mechanisms involved in duplicating the human genome is critical to further development of therapeutic drugs.

The "ORC cycle": a novel pathway for regulating eukaryotic DNA replication

Ghosh, Noguchi, Radichev, Vassilev, Zhang

We wish (1) to determine how and when cells regulate the activity of the "origin recognition complex" (ORC), a six-subunit complex whose interaction with the genome determines where DNA replication begins, and (2) to discover communication links between DNA replication and other proteins that regulate cell growth and development. Such links are needed to maintain homeostasis by telling the cell when replication has begun, when it is complete, and when it is aberrant.

In 2000, we reported the first clear evidence that ORC activity in mammalian cells is regulated by cell cycle changes in the affinity of the largest subunit (Orc1) for chromatin, which has since been confirmed by evidence from several laboratories. Further work has shown that mammalian Orc1 is selectively ubiquitinated and phosphorylated during the S- to M-phase transition while ORC subunits 2 to 5, which constitute a stable core complex, remain tightly bound to chromatin throughout cell division (see Figure 17.1). In G1-phase cells, only ORC(1-5) complexes are bound to chromatin; ORC(2-5) complexes can be assembled in the nucleus but do not bind to chromatin. In human tumor cells, Orc1 is selectively degraded by a ubiquitination-dependent mechanism. This selective destabilization of Orc1 requires the site-specific binding of CcnA to Orc1, which then recruits Cdk1 during S-phase. Presumably, loss of Orc1 inactivates ORC, and the ORC(2-5) complex is released from chromatin during mitosis. Cdk1 is inactive until G2-phase, when Cdc25 phosphatase activates it and thus allows Cdk1-CcnA or Cdk1-CcnB to phosphorylate Orc1 and thereby prevent it from binding to chromatin until mitosis is completed and a nuclear membrane has formed. During mitosis, neither Orc1 nor the stable ORC(2-5) complex is bound to chromatin. Binding of ORC to chromatin can occur only when Orc1 has been dephosphorylated as cells exit from metaphase, although binding can be driven to occur as early as anaphase by increasing the cellular level of Orc1. Thus, it appears that ORC-chromatin sites are disbanded as pre-replication complexes are activated to initiate DNA synthesis and then reassembled on chromosomes as cells transit from M- to G1-phase. This means that cells have the opportunity to reprogram their initiation sites each time they divide, changing the number of sites as well as their locations. Binding of Orc1 to replication origins in chromatin and its ability to activate those origins is facilitated by the single BAH domain in Orc1. Orc1 is the only ORC subunit that contains a BAH domain, which is involved in protein-protein interactions. However, the Orc1 BAH domain does not appear to affect either ORC assembly or the selective degradation of Orc1 during S-phase. Instead, it facilitates binding of ORC to chromatin during the M- to G1-phase transition.

FIGURE 17.1

Regulation of ORC activity in mammalian cells. All six ORC subunits (shaded cylinders) are tightly bound to chromatin during G1-phase to provide active ORC:chromatin sites for initiation of preRC assembly. DNA synthesis does not begin until preRCs are activated by CcnE-Cdk2, a protein kinase whose activity is inhibited during G1-phase by the CDK-specific inhibitor p27/Kip1. During S-phase, Orc1 is specifically bound to CcnA-Cdk1. The affinity of Orc1 for chromatin appears to selectively reduced and is targeted for ubiquitination by SCFSkp2. Given that SCFSkp2-dependent degradation of p27 is essential for cell cycle progression, CcnE-Cdk2 activation is accompanied by Orc1 inactivation, resulting in the initiation of S-phase with concomitant suppression of reinitiation of preRC assembly. In M-phase cells Orc1 is hyperphosphorylated, associated with CcnA-Cdk1, and does not bind tightly to chromatin. During the M to G1 phase transition, CcnA is degraded, Cdk1 is inactivated, and Orc1 is dephosphorylated and bound to chromatin.

Transient ectopic expression of unmodified Orc1 rapidly induces p53-independent apoptosis. Remarkably, co-expression of Orc1 with Orc2, the only ORC subunit that did not induce apoptosis, prevented Orc1 induction of apoptosis, suggesting that assembly of ORC:chromatin sites during the M- to G1-phase transition of cell division neutralizes the toxic effects of individual subunits. Apoptosis was also suppressed by either the addition of a single ubiquitin to Orc1 or the quasi-hyperphosphorylation of Orc1 (conversion of S/T to D at Cdk phosphorylation sites), suggesting that failure to carry out these modifications during cell proliferation would induce apoptosis. Moreover, these modifications caused Orc1 to localize to the cytoplasm where it cannot participate in the assembly of ORC:chromatin sites, thus confirming that these Orc1 modifications would indeed suppress ORC activity during the S- and G2/M-phases.

Various manifestations of this ORC cycle appear in yeast, frogs, and flies. In frogs, we have shown that the entire XlORC is released from somatic cell chromatin following assembly of pre-replication complexes in G1-phase. In flies, others have shown that DmOrc1 is selectively degraded during mitosis. In yeast, still others have shown that Orc2 and Orc6 are phosphorylated when cells enter S-phase and that this event reduces ORC activity. Thus, there is a universal control point in eukaryotic cell division cycles: the same cyclin-dependent protein kinase that regulates the onset of mitosis (Cdk1) also prevents premature assembly of functional ORC/chromatin sites until mitosis is complete and a nuclear membrane is present.

Tead2 and Dkkl1: two closely linked genes that are differentially expressed during mammalian development

Kaneko, Kohn, Yagi, Ullah, Zhang; in collaboration with Liu, Maraia

Our goal is to identify the roles of Tead:Yap65 transcription factor complexes in the development and differentiation of embryonic stem (ES) and trophoblast stem (TS) cells. ES and TS cells appear at the blastocyst stage in mammalian development, a few days after fertilization when cell differentiation first occurs. ES cells give rise to all embryonic cell lineages, but TS cells give rise to the placental cell lineages, including the trophoblast giant cells that undergo endoreduplication and appear to be involved in implantation of the blastocyst in the uterine endometrium.

Mammals express four highly conserved Tead(TEF) transcription factors that recognize a canonical M-CAT motif (5′-CATTCCT-3′) found in promoters specific for transcription in muscle as well as similar motifs found in SV40 and polyomavirus enhancers. Tead-1(TEF-1) is required for mouse cardiac development by day 10 of embryonic development and for gene expression in cardiac muscle cells. Tead-4(TEF-3) appears to play a specific role in activating skeletal muscle genes. Tead-3(TEF-5) is expressed primarily in the placenta and in cardiac muscle. In adult mice, Tead-2(TEF-4) is expressed strongly in heart and lung tissues and in the granulosa cells of the ovary and weakly in several other tissues. However, we showed that Tead2 is the only Tead gene expressed in mouse embryos during the first seven days of development, suggesting that it plays a unique role at the beginning of mammalian development by allowing pre-implantation mouse embryos to use Tead-dependent promoters and enhancers that we had shown previously to function in pre-implantation embryos and embryonic stem cells.

While searching for Tead2-regulatory elements, we discovered a novel single-copy gene that is now called Dickkopf-like 1 (Dkkl1) (formerly called Soggy [Sgy]) located 3.8 kb upstream of the Tead2 mRNA start site and transcribed in the opposite direction. Dkkl1 is unique to mammals; it is always closely linked to Tead2 and expressed in very few cell types. It is related to a group of secreted proteins that are antagonists of Wingless (Wnt) signal transduction pathways. Thus, Dkkl1- and Tead2-regulatory elements are proximate to one another and provide an example of two closely spaced, divergently transcribed genes. They also provide a unique paradigm for differential regulation of gene expression during mammalian development.

Both Dkkl1 and Tead2 are among the first genes to be expressed at the beginning of mouse development. In pre-implantation embryos, Dkkl1 is selectively expressed in trophoblast stem cells and concomitantly repressed in embryonic stem cells (see Figure 17.2). Thus, Dkkl1 appears to be required in the placental lineage, where it appears in the trophectoderm and eventually in the trophoblast giant cells that are involved in implantation, but it is toxic to the embryonic lineage. In adult mammals, Dkkl1 is expressed predominantly, although not exclusively, during formation of the male germ cells where it eventually localizes in the acrosome of mature sperm. Following capacitation, some Dkkl1 protein migrates to the surface of the sperm where it may be involved in fertilization. Thus, Dkkl1 is involved in two seemingly unrelated functions: production of sperm and production of trophoblast cells and their derivatives.

Publications Related to Other Work

COLLABORATORS

Chengyu Liu, PhD, Transgenic Facility, NHLBI, Bethesda, MD
Richard Maraia, MD, Laboratory of Molecular Growth Regulation, NICHD, Bethesda, MD

For further information, contact depamphm@mail.nih.gov or visit http://depamphilislab.nichd.nih.gov.