DNA repair

Right on target with ubiquitin

Article metrics

Cellular DNA-repair mechanisms prevent mutations from accumulating, thereby averting defects in cell function. A molecule best known for its role in protein degradation is now shown to have a specific task in DNA repair.

The activity of many of the proteins in our cells depends on the chemical 'labels' that are attached to them. A commonly used label is the ubiquitin molecule, which regulates numerous cellular processes1. Many of ubiquitin's regulatory functions reflect its role as a tag that singles out proteins to be degraded1. Yet 'ubiquitination' can also signal other fates2. The idea that cells can interpret ubiquitin signals in diverse ways was first suggested more than a decade ago, when a protein called Rad6 — which is needed for cells to repair damaged DNA3 — was found to be an enzyme that helps to join ubiquitin to target proteins4. Despite this early hint, further evidence of a role for ubiquitin in protecting DNA was elusive: the relevant target proteins, and the consequences of their modification, remained unknown. But, on page 135 of this issue, Hoege and co-workers5 identify the first functionally relevant target of ubiquitination during DNA repair — a protein called PCNA.

DNA is highly susceptible to environmental insults that can alter its sequence (causing mutations), or prevent it from being copied altogether (causing cell death)6. For instance, DNA damage caused by chemicals or ultraviolet light can block the progress of the enzymes that copy DNA — DNA polymerases — creating gaps in one of the two strands of a newly produced DNA molecule. Ubiquitination is a positive signal in a pathway that permits DNA replication to be completed despite such damage7. This pathway, which requires the Rad6 protein, does not remove the original lesions, but instead uses 'post-replicative' DNA synthesis to fill in the gaps, using an undamaged strand as a template. The synthesis can follow either of two routes, one error-prone and the other error-free. The error-free mechanism is especially important — if it fails, the error-prone mechanism takes over, generating mutations in the newly made DNA. Post-replicative repair is the least well understood of the several DNA-repair mechanisms in higher organisms6.

The main type of error-free post-replicative repair requires five enzymes, including Rad6, that conjugate ubiquitin to target proteins (Fig. 1)7. The pathway also relies on a specialized signal in which several ubiquitins are chained together in a particular way8. These properties bespeak an intimate involvement of ubiquitin in DNA repair, but several questions remained unanswered until now. Which proteins are modified with ubiquitin by the dedicated conjugating enzymes? Which polymerase (or polymerases) carries out the bulk of error-free post-replicative synthesis? And what is the consequence of labelling the target proteins with ubiquitin? In providing an answer to the first question — that PCNA is ubiquitinated during post-replicative repair — the work of Hoege et al.5 sets the stage for answering the second and third.

Figure 1: Connecting DNA repair and ubiquitin, through the PCNA protein.

The PCNA trimer, shown at the left, encircles DNA and binds to DNA-replicating enzymes (polymerases). Hoege et al.5 have found that one complex of ubiquitin-conjugating proteins (Rad18 and Rad6) attaches a single ubiquitin (Ub) to a specific lysine amino acid in PCNA. A second conjugating complex (consisting of Rad5, Mms2 and Ubc13) extends a polyubiquitin chain from this first ubiquitin. The modified PCNA then promotes error-free post-replicative DNA repair. Modification of this same lysine amino acid by SUMO, another member of the ubiquitin family, depends on a distinct conjugating factor (Ubc9) and inhibits such DNA repair.

PCNA was already known to work with one DNA polymerase in DNA replication9, and with others in DNA repair. How did Hoege et al. discover that PCNA is also a target for ubiquitination? As is often the case, serendipity was important. The researchers actually set out to study SUMO, one of several ubiquitin-like proteins (the name stands for 'small ubiquitin-related modifier') that act through a similar conjugation mechanism10. Hoping to gain insight into SUMO-mediated signalling, the authors purified 'sumoylated' target proteins from yeast cells, and found that PCNA was among them.

While confirming this finding, Hoege et al. also noticed other modified forms of PCNA, which proved to be attached to the special chain of ubiquitins that is the signature of error-free repair. This 'polyubiquitination', but not sumoylation, required the Rad6 pathway to be active and was induced by DNA damage. Moreover, DNA repair was inhibited when the authors mutated the lysine amino acid in PCNA to which polyubiquitin becomes attached, and genetic analysis showed that this effect resulted from blocking the conjugation of PCNA to ubiquitin. So PCNA is truly a repair-relevant substrate for ubiquitination. But the effect on DNA repair of mutating PCNA's ubiquitin-attachment site is weaker than the effect of blocking ubiquitin conjugation altogether, implying that further targets of polyubiquitination remain to be discovered.

So the polyubiquitination of PCNA is needed for DNA repair. What about the sumoylation of this protein? Ubiquitin and SUMO are attached to target proteins by different conjugating factors. Nonetheless, they can modify the same lysine residue of PCNA. Hoege et al. present several lines of evidence that suggest that occupation of this lysine by SUMO inhibits DNA repair. Such antagonism between ubiquitin and SUMO was known in protein degradation11, but it seems that the strategy may apply more broadly.

With a known substrate in hand, Hoege and co-workers were also able to gain insight into the specific roles of the various ubiquitin-conjugation enzymes. Apparently, a single ubiquitin attached to the crucial lysine of PCNA by one conjugating complex serves as the starting point for extension of the chain by a different complex. This persuasive model agrees with the order of events suggested by earlier genetic analyses and is consistent with several known protein– protein interactions7,12. The production of a polyubiquitin chain by the sequential action of different conjugating factors is a new observation; it will be interesting to see which other ubiquitin-conjugating systems use this strategy.

PCNA had been implicated previously in post-replicative repair13 and is also involved in other repair pathways9. The results of Hoege et al. confirm the relevance of PCNA to Rad6-mediated repair, and lend weight to the idea that the DNA polymerase with which PCNA works in DNA replication is also intimately involved in error-free repair14. PCNA assembles into a trimeric ring that encircles DNA and promotes faithful and efficient replication by DNA polymerases. It also provides a platform for recruitment of regulatory factors9. The lysine residue modified by SUMO or polyubiquitin lies at the edge of the PCNA ring. One can easily imagine the polyubiquitin chain recruiting new molecules to a stalled polymerase, promoting its removal, relocalization or reprogramming. So the new results are the first to suggest how repair and replication could be integrated at the molecular level.

The results also raise new questions — for example, is the switch between sumoylation and ubiquitination controlled through attachment exclusively, or is the removal of SUMO and ubiquitin regulated too? And does attachment of a single ubiquitin (rather than a chain) to PCNA promote a different mode of repair, as the authors propose5? Finally, two of the ubiquitin-conjugating factors involved have additional biochemical activities: Rad18 binds single-stranded DNA and Rad5 may unwind DNA7. Whether these properties are relevant in interpreting the ubiquitin signal, as well as in generating it, remains to be seen.


  1. 1

    Hershko, A. & Ciechanover, A. Annu. Rev. Biochem. 67, 425–479 (1998).

  2. 2

    Pickart, C. M. Trends Biochem. Sci. 25, 544–548 (2000).

  3. 3

    Reynolds, P., Weber, S. & Prakash, L. Proc. Natl Acad. Sci. USA 82, 168–172 (1985).

  4. 4

    Jentsch, S., McGrath, J. P. & Varshavsky, A. Nature 329, 131–134 (1987).

  5. 5

    Hoege, C., Pfander, B., Moldovan, G.-L, Pyrowolakis, G. & Jentsch, S. Nature 419, 135–141 (2002).

  6. 6

    Hoeijmakers, J. H. J. Nature 411, 366–374 (2001).

  7. 7

    Broomfield, S., Hryciw, T. & Xiao, W. Mutat. Res. 486, 167–184 (2001).

  8. 8

    Spence, J., Sadis, S., Haas, A. L. & Finley, D. Mol. Cell. Biol. 15, 1265–1273 (1995).

  9. 9

    Warbrick, E. BioEssays 22, 997–1006 (2000).

  10. 10

    Hochstrasser, M. Nature Cell Biol. 2, E153–E157 (2000).

  11. 11

    Desterro, J. M., Rodriguez, M. S. & Hay, R. T. Mol. Cell 2, 233–239 (1998).

  12. 12

    Ulrich, H. D. & Jentsch, S. EMBO J. 19, 3388–3397 (2000).

  13. 13

    Torres-Ramos, C., Yoder, B. L., Bugers, P. M. J., Prakash, S. & Prakash, L. Proc. Natl Acad. Sci. USA 93, 9676–9681 (1996).

  14. 14

    Torres-Ramos, C. A., Prakash, S. & Prakash, L. J. Biol. Chem. 272, 25445–25448 (1997).

Download references

Author information

Correspondence to Cecile M. Pickart.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Pickart, C. Right on target with ubiquitin. Nature 419, 120–121 (2002) doi:10.1038/419120a

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.