Control of replication is necessary in order to maintain a stable genome that can be segregated properly into two daughter cells at mitosis. Each part of the genome must be copied precisely once during S phase, with no omissions or duplications. Furthermore this copying process has to be error free to avoid mutation. The chromatin structure of a chromosome also contains epigenetic information that must be duplicated in this crucial cell cycle stage. The diploid human genome contains 6 billion base pairs of DNA in 46 chromosomes so this is a formidable task!
The complex processes of DNA replication are under many levels of cellular control. The first control is the decision of where to start the copying process. In higher eukaryotes DNA synthesis is initiated at discreet loci, called origins, which are recognised by the multiprotein origin recognition complex ORC. In order to ensure that each origin is only activated once during the cell cycle cells have a control system known as licensing. In G1 phase before replication starts origins are licensed by the action of, among other proteins, Cdc6 and Cdt1, which trigger the association of replication initiation proteins, such as the MCMs, with ORC. This is however not sufficient to start replication. Firing of these licensed origins requires the transition to S phase and the activation of S-phase kinases, including Cdc7/Dbf4. The activation of these kinases occurs at the same time as the inactivation of the licensing proteins, preventing the re-licensing of origins that have already fired (for a recent review, see ref 1). Interestingly, the order in which origins fire is strictly regulated; some regions of the genome are always copied early and some late. This control may be mediated by subnuclear compartmentalisation and chromatin structure, at least in yeast2.
Once an origin fires, the two strands of DNA are separated by a helicase moving away from the origin, which in human cells is probably the MCM2-7 complex. DNA synthesis on the separated template strands is initiated by primase, which makes a short RNA primer that can be extended by the DNA polymerase pola. The multisubunit replicative polymerases pold and pole are responsible for copying the majority of the genome. Of the two strands, one (the leading strand) is copied in a continuous manner and the other (the lagging strand) in a semi-discontinuous manner. Synthesis of the lagging strand is therefore more complex as the short Okazaki fragments need to be joined correctly to obtain a completed chromosome. This process involves removal of RNA primers and overhanging DNA and ligation (by Fen1 and ligase I respectively) (for detailed review see refs 3 and 4). Under conditions of stress, such as depleted nucleotides or DNA damage, the replicative polymerases may not be able to complete replication as damaged DNA cannot be used as a template by the highly accurate replicative polymerases. In this case one of the “error prone” polymerases, such as polh or polz can be used to overcome the block and allow replication to continue (reviewed in ref 5). Although the replicative polymerases are accurate, and have proofreading activity, errors do still occasionally occur. A large proportion of these are removed after replication by the mismatch repair machinery (requiring Msh2 and Msh6), which detects and excises bases incorrectly incorporated into the newly synthesised strands6.
A key component of the replication machinery, that seems to regulate all of these steps, is the homotrimeric sliding clamp PCNA. This protein topologically encircles the DNA at the site of synthesis and it is used by the replicative polymerases as a processivity factor, ensuring that they can synthesize large stretches of DNA without falling off their template. PCNA has been shown to also interact with proteins involved in all the steps of replication described above, including Cdc6, Cdt1, pola, Fen1, Ligase I, pold, pole, polh, polz, Msh2, and Msh6.(7) How can all these interactions with a single molecule be properly regulated? The answer to this question is still not at all clear. Visualisation of any of these proteins in the cell nucleus during S phase by various microscopic techniques has shown that the active processes of replication occur within discreet nuclear foci, known as replication factories.(8) The figure below shows a Z stack of consecutive optical sections through an S phase nucleus stained for PCNA. Each dot is a factory, where multiple replication forks are active. Similar patterns have been demonstrated for most known replication proteins, indeed co-localisation with PCNA is often used as a cellular marker for proteins involved in replication 9,10. These factories certainly represent extremely elevated local concentrations of active replication forks and key replication proteins, but the mechanisms regulating traffic within a factory still remain to be determined.
Figure 1. Replication factories
Figure 1 – Replication factories in an MRC5 human fibroblast cell visualised by indirect immunofluorescence of PCNA (primary antibody PC10 – Abcam). Numbers in white represent the depth of the Z section in mm. Cells were fixed in paraformaldehyde after triton extraction to remove soluble proteins, then methanol treated to reveal the PC10 epitope. Images were acquired on a Zeiss LSM510 confocal microscope.
Accurate completion of DNA synthesis is not sufficient for proper transmission of all inherited information to daughter cells. The chromosomes also contain epigenetic information, encoded within DNA methylation patterns, histone modifications and chromatin structure. This information must also be copied during S phase.(11) The maintenance DNA methyl transferases are found in replication factories and interact with PCNA, suggesting that the copying of DNA methylation status also occurs at the replication fork.(9) Furthermore the essential chromatin assembly factor CAF-1 has similar localisation, and it deposits histones H3 and H4 at replication sites. (12,13) The coupling of DNA synthesis and chromatin assembly is crucial as down regulating CAF-1 leads to genomic instability and cell death. How epigenetic information is transmitted from parental chromatin to the two daughters is not entirely clear. For example, CAF-1 assembles nucleosomes using the newly synthesised histones that are specifically acetylated on lysines 5 and 12 of histone H4. These acetylation marks are altered as the cell cycle progresses in a locus specific manner.(14) The histones that made up the parental chromatin are incorporated into the newly synthesized fibre, and this may be the mechanism by which cells ensure that histone codes are inherited, but more work needs to be done to demonstrate this.(15,16) After nucleosome assembly at the fork, a process of chromatin maturation involving chromatin remodelling enzymes such as the WICH complex occurs to produce completely finalised duplicated chromosomes. The enzymes responsible for this process also seem to be present in the replication foci.(17)
In summary, replication of the genome and the epigenome is a vastly complicated procedure involving hundreds of different enzymes that must all be strictly regulated to work co-ordinately. The cell achieves this by restricting replication both temporally and spatially in replication factories. These factories contain all the activities necessary for the duplication of the genome, but how they function on a molecular level is a question that will surely be the subject of much research in the near future.
- Lehmann AR. Replication of damaged DNA in mammalian cells: new solutions to an old problem. Mutat Res. 2002 Nov 30;509(1-2):23-34.
- Jiricny J. The multifaceted mismatch-repair system. Nat Rev Mol Cell Biol. 2006 May;7(5):335-46.
- Maga G, Hubscher U. Proliferating cell nuclear antigen (PCNA): a dancer with many partners. J Cell Sci. 2003 Aug 1;116(Pt 15):3051-60.
- Cook PR. The organization of replication and transcription. Science. 1999 Jun 11;284(5421):1790-5.
- Leonhardt H, Rahn HP, Weinzierl P, Sporbert A, Cremer T, Zink D, Cardoso MC. Dynamics of DNA replication factories in living cells. J Cell Biol. 2000 Apr 17;149(2):271-80.
- Bienko M, Green CM, Crosetto N, Rudolf F, Zapart G, Coull B, Kannouche P, Wider G, Peter M, Lehmann AR, Hofmann K, Dikic I. Ubiquitin-binding domains in Y-family polymerases regulate translesion synthesis. Science. 2005 Dec 16;310(5755):1821-4.
- McNairn AJ, Gilbert DM. Epigenomic replication: linking epigenetics to DNA replication. Bioessays. 2003 Jul;25(7):647-56.
- Krude T. Chromatin assembly factor 1 (CAF-1) colocalizes with replication foci in HeLa cell nuclei. Exp Cell Res. 1995 Oct;220(2):304-11.
- Verreault A, Kaufman PD, Kobayashi R, Stillman B. Nucleosome assembly by a complex of CAF-1 and acetylated histones H3/H4. Cell. 1996 Oct 4;87(1):95-104.
- Taddei A, Roche D, Sibarita JB, Turner BM, Almouzni G. Duplication and maintenance of heterochromatin domains. J Cell Biol. 1999 Dec 13;147(6):1153-66.
- Benson LJ, Gu Y, Yakovleva T, Tong K, Barrows C, Strack CL, Cook RG, Mizzen CA, Annunziato AT. Modifications of H3 and H4 during chromatin replication, nucleosome assembly, and histone exchange. J Biol Chem. 2006 Apr 7;281(14):9287-96.
- Tagami H, Ray-Gallet D, Almouzni G, Nakatani Y. Histone H3.1 and H3.3 complexes mediate nucleosome assembly pathways dependent or independent of DNA synthesis. Cell. 2004 Jan 9;116(1):51-61.
- Poot RA, Bozhenok L, van den Berg DL, Hawkes N, Varga-Weisz PD. Chromatin remodeling by WSTF-ISWI at the replication site: opening a window of opportunity for epigenetic inheritance? Cell Cycle. 2005 Apr;4(4):543-6.