DNA replication

Strand separation unravelled

Article metrics

The DNA double helix must be separated into single strands to be duplicated. A structure of the Mcm2–7 helicase enzyme responsible for this activity yields unprecedented insight into how the process is initiated. See Article p.186

The successful replication of double-stranded DNA, an essential part of cell division, depends on a helicase enzyme that separates the two component strands. Although simple helicases have been extensively studied1, much less is known about the complex replicative helicases found in eukaryotes (the group of organisms that includes animals, plants and fungi). But that is about to change. On page 186 of this issue, Li et al.2 capitalize on advances in cryo-electron microscopy3 to resolve the structure of a eukaryotic helicase, Mcm2–7, to a near-atomic resolution of 3.8 ångstroms — around five times higher than the best Mcm2–7 structure reported so far4. Combined with previous studies, this structure indicates how a key step in DNA replication occurs: the initial 'melting' of double-stranded DNA into single strands.

Mcm2–7 has a central role in eukaryotic DNA replication. Like similar helicases from bacteria, archaea and viruses, it unwinds double-stranded DNA (dsDNA) by binding one strand in its central channel, excluding the other. Energy, provided by the enzyme's ability to hydrolyse ATP molecules, enables the complex to translocate along the bound DNA, resulting in unwinding of the complementary strand1. But the doughnut-shaped Mcm2–7 is structurally and functionally different from other helicases, because it is the only known hexameric helicase to be derived from six different subunits (Mcm2 to Mcm7) instead of from six copies of the same subunit. This feature has allowed portions of the complex to evolve extra, specialized functions that are thought5 to be crucial to the enzyme's ability to load onto DNA and to activate its unwinding activity — two landmark regulatory events during DNA replication.

Two structures containing Mcm2–7 have been described previously. One represents the CMG complex6,7, which is active during the phase of DNA replication known as elongation, when complementary DNA is synthesized for each existing strand. The complex contains one Mcm2–7 hexamer and two other essential replication factors that activate the enzyme's DNA unwinding ability. By contrast, the second structure8,9 is an inactive form of the enzyme, which has been isolated from cells before they replicate. This structure contains two Mcm2–7 hexamers in a head-to-head orientation, enclosing dsDNA in the central channel. Helicase structures such as this Mcm2–7 double hexamer (Mcm2–7 DH) are rare, and so its purpose has been a cause for debate.

One reasonable conjecture is that the Mcm2–7 DH participates in DNA melting. Whereas DNA unwinding enlarges a pre-existing single-stranded DNA (ssDNA) region during elongation, DNA melting, which is an earlier process, initiates replication by locally transforming dsDNA into ssDNA. Local melting provides a site for the subsequent assembly of a DNA replication fork — the full complement of proteins that enable duplication of the genetic material10. Although melting has been well studied in bacteria, little is known about how it occurs in eukaryotes10. Li and colleagues' structure, when combined with other data, is highly consistent with a role for the Mcm2–7 DH in DNA melting, for several reasons.

First, both the current study and a previous one4 demonstrate that the two Mcm2–7 hexamers in the Mcm2–7 DH are offset along the long vertical axis of the hexamer, at a 14° tilt relative to one another (Fig. 1a). This offset restricts the dimensions of the central channel (Fig. 1b). Although DNA is not visible in the authors' structure, these data suggest that dsDNA will be kinked at the interface between the two hexamers. Sharp DNA bending is known to cause local DNA melting11, and may contribute to the unwinding of dsDNA during transcription12. Thus, a DNA kink between the two Mcm2–7 hexamers could serve to initiate DNA melting.

Figure 1: Structure of a strand separator.

a, The Mcm2–7 helicase is a doughnut-shaped enzyme composed of six different subunits (individual subunits not shown) that is vital for DNA replication5. Li et al.2 resolve a structure that contains two copies of this hexamer, which they suggest is involved in the DNA melting process that converts double-stranded DNA (dsDNA) into single strands to initiate DNA replication. In the authors' structure, the two hexamers are tilted at a 14° angle relative to one another. b, This conformation forms a narrow central channel between the two hexamers through which dsDNA can pass, together with two exit channels. c, dsDNA pumped through the central channel might be extruded through the exit channels as single-stranded DNA 'rabbit ears', which can then be replicated.

Second, although helicases normally interact productively only with ssDNA, a specific form of the Mcm complex (Mcm467) has been shown to bind to and translocate along dsDNA13. This is consistent with a potential role for Mcm2–7 in manipulating dsDNA during melting. Finally, unlike bacterial hexameric helicases, some viral replicative helicases, such as papillomavirus E1 and simian virus-40 (SV40) large T-antigen, initially form dsDNA-containing DHs that resemble the Mcm2–7 DH (refs 14,15). These structures locally melt DNA and then uncouple into single hexamers to unwind DNA during elongation.

How might dsDNA melting occur? Electron microscopy indicates that the SV40 large T-antigen DH can act as a pump13, in which dsDNA enters each hexamer from flanking regions and ssDNA is extruded in 'rabbit ear' structures at the interface between them14. Consistent with such a mechanism in eukaryotes, the misalignment of the two hexamers in the Mcm2–7 DH creates two exit channels at the hexamer interface through which rabbit ears might be extruded (Fig. 1c). Thus, the Mcm2–7 DH might melt DNA in a manner analogous to melting on SV40 large T-antigen, with local unwinding of the bent DNA forming a highly flexible hinge to facilitate ssDNA extrusion. Such a model had been proposed to explain Mcm2–7 DNA unwinding during elongation13. Because the Mcm2–7 DH seems to be enzymatically inactive, further research will be needed to identify the factors required to activate the DH for melting, as well as to determine how the individual Mcm2–7 hexamers physically uncouple and are remodelled into the ssDNA-bound form needed for elongation.

Given the technical advances in cryo-electron microscopy, a flood of high-resolution structures should become available in the near future. However, such structures provide only a static glimpse of the target protein, a particularly limiting problem for the study of dynamic processes such as DNA replication. Because Mcm2–7 is only one of many molecular motors involved in DNA replication, understanding the dynamic nature of their interactions is essential for a complete understanding of DNA replication. To this end, single-molecule studies using reconstituted eukaryotic replication systems7,16 have begun to shed much-needed light on the dynamics of this process. Together, these varied experimental approaches should yield a holistic understanding of the vital process of DNA replication.Footnote 1


  1. 1.

    See all news & views


  1. 1

    Enemark, E. J. & Joshua-Tor, L. Curr. Opin. Struct. Biol. 18, 243–257 (2008).

  2. 2

    Li, N et al. Nature 524, 186–191 (2015).

  3. 3

    Kühlbrandt, W. eLife 3, e03678 (2014).

  4. 4

    Sun, J. et al. Genes Dev. 28, 2291–2303 (2014).

  5. 5

    Bochman, M. L. & Schwacha, A. Microbiol. Mol. Biol. Rev. 73, 652–683 (2009).

  6. 6

    Ilves, I., Petojevic, T., Pesavento, J. J. & Botchan, M. R. Mol. Cell 37, 247–258 (2010).

  7. 7

    Fu, Y. V. et al. Cell 146, 931–941 (2011).

  8. 8

    Evrin, C. et al. Proc. Natl Acad. Sci. USA 106, 20240–20245 (2009).

  9. 9

    Remus, D. et al. Cell 139, 719–730 (2009).

  10. 10

    Gai, D., Chang, Y. P. & Chen, X. S. Curr. Opin. Struct. Biol. 20, 756–762 (2010).

  11. 11

    Shin, J., Lee, O.-C. & Sung, W. J. Chem. Phys. 142, 155101 (2015).

  12. 12

    Tang, G.-Q. & Patel, S. S. Biochemistry 45, 4936–4946 (2006).

  13. 13

    Kaplan, D. L. & O'Donnell, M. Mol. Cell 15, 453–465 (2004).

  14. 14

    Schuck, S. & Stenlund, A. Mol. Cell 43, 776–787 (2011).

  15. 15

    Wessel, R., Schweizer, J. & Stahl, H. J. Virol. 66, 804–815 (1992).

  16. 16

    Ticau, S., Friedman, L. J., Ivica, N. A., Gelles, J. & Bell, S. P. Cell 161, 513–525 (2015).

Download references

Author information

Correspondence to Matthew L. Bochman or Anthony Schwacha.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Bochman, M., Schwacha, A. Strand separation unravelled. Nature 524, 166–167 (2015) doi:10.1038/nature14643

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.