Cells replicate half of their genome as short fragments that are put together later on. The way in which this process is linked to the formation of DNA–protein complexes called nucleosomes is now becoming clearer. See Article p.434
In a typical human cell, about two metres of DNA is wrapped around proteins to form nucleosomes, the basic units of chromatin. When this DNA is duplicated, the nucleosomes that lie ahead of the replication machinery are displaced, or disassembled into their component histone proteins. And after the machinery has passed by, the nucleosomes are reassembled on the two resulting 'daughter' strands, which requires the addition of new histones1.To make things more complicated, one of the two DNA strands can be replicated only in short pieces — called Okazaki fragments — that are joined together afterwards, and so its replication lags behind that of the other strand (Fig. 1). This elaborate mechanism is needed because DNA synthesis is unidirectional, whereas the sequences of the two DNA strands run in opposite directions. However, the links between DNA replication and nucleosome dynamics have previously been studied without distinguishing between the two strands. On page 434 of this issue, Smith and Whitehouse2 uncover the interplay between nucleosome assembly and the synthesis of the lagging strand.
The authors2 analysed Okazaki fragments from baker's yeast, a unicellular fungus commonly used as a model for eukaryotic cells — those found in organisms such as animals, plants and fungi. Smith and Whitehouse took advantage of an available large catalogue of defined yeast mutants, together with well annotated genomic information that included recently defined nucleosome positions3. The researchers developed an elegant method for purifying Okazaki fragments from yeast, then sequenced the fragments and mapped them on the yeast genome. The result is the first high-resolution, genome-wide analysis of Okazaki fragments from live eukaryotic cells. The analysis is a major achievement, as our knowledge of lagging-strand synthesis has been based largely on model studies4 of the replication of non-chromosomal DNA molecules such as viral DNA.
Smith and Whitehouse discovered that, strikingly, the Okazaki fragments' size and location in the genome depend on where the nucleosomes are located. More specifically, the ends of the Okazaki fragments fall near the nucleosome's dyad axis, which corresponds to the midpoint of the DNA segment wrapped around the chromatin unit. This result contrasts with a common assumption that, during lagging-strand replication, a nucleosome would form as soon as the Okazaki fragment was long enough, and then the next fragment would be synthesized. This assumption tallies with the observation that the average length of Okazaki fragments is similar to that of the DNA wound around a nucleosome.
The authors2 propose an alternative model in which the presence of a nascent nucleosome acts as a roadblock to an advancing polymerase enzyme (Pol δ), which is synthesizing the lagging strand. According to their model, Pol δ and other associated enzymes can invade the nascent nucleosome up to its dyad axis, but then the nucleosome mediates the dissociation of all of these enzymes from the DNA — by doing so, the nucleosome determines the length of the Okazaki fragment.
To test the model, Smith and Whitehouse used yeast strains in which the physiological balance between polymerases and nucleosomes was altered, either because the polymerases were less active or because nucleosome assembly was impaired. For example, they observed2 that the Okazaki fragments were longer in mutant strains defective in CAF-1, a protein complex known5 to act as a histone chaperone, which promotes histone deposition on newly synthesized DNA. Therefore, the tight balance between the capacity to form a new nucleosome on the lagging strand and the polymerase's ability to advance is what sets the length of the Okazaki fragment. These findings also suggest that the efficiency of the replication machinery's progression may depend on where the nucleosome forms.
The authors' work provides the basis for research on how other factors involved in histone dynamics and nucleosome movement on DNA — such as histone chaperones5 and chromatin remodeller enzymes — can affect Okazaki-fragment size and, overall, DNA replication. Moreover, additional proteins involved in chromatin assembly could be identified by screening the available collection of yeast mutants for changes in Okazaki-fragment length. And the technique used for mapping Okazaki fragments offers a means to identify the genomic sites at which DNA replication starts.
It is tempting to postulate that the inherent asymmetry of nucleosome dynamics during DNA replication could contribute to a mechanism for asymmetric cell division, a process by which two different cell types are generated. This hypothesis is supported by the finding6 that CAF-1 has a role in asymmetric cell division in the worm Caenorhabditis elegans. In addition, histones in nucleosomes exist in different forms associated with specific genomic regions, and can be 'marked' with various chemical modifications that reflect functional states, such as activation or repression of gene expression. Whereas some of these modifications are transient, others may be inherited during cell division in an epigenetic fashion — that is, independently of the DNA sequence. But to understand how the chemical modifications of nucleosomes can be inherited during cell division, we must first determine how nucleosomal organization is reproduced during DNA replication. So Smith and Whitehouse's study paves the way for a deeper exploration of the intimate relationship between genetics and epigenetics.
Saha, A., Wittmeyer, J. & Cairns, B. R. Nature Rev. Mol. Cell Biol. 7, 437–447 (2006).
Smith, D. J. & Whitehouse, I. Nature 483, 434–438 (2012).
Jiang, C. & Pugh, B. F. Genome Biol. 10, R109 (2009).
Anderson, S. & DePamphilis, M. L. J. Biol. Chem. 254, 11495–11504 (1979).
Corpet, A. & Almouzni, G. Trends Cell Biol. 19, 29–41 (2009).
Nakano, S., Stillman, B. & Horvitz, H. R. Cell 147, 1525–1536 (2011).
About this article