Taking the measure

Telomeres — the tips of chromosomes — need to be preserved, and this involves replenishing telomeric DNA when it has been eroded. But telomeres must not become too long, and one aspect of length control is now revealed.

In every organism, maintaining the integrity of the genome is a crucial endeavour. One aspect of genome maintenance involves protecting telomeres, the natural ends of linear chromosomes. This task is achieved by a suite of specialized protein complexes, which are anchored to chromosome ends through their association with further proteins that bind directly to telomeric DNA. The resulting structure prevents events that would be catastrophic for the genome, such as the loss of terminal DNA sequences or end-to-end chromosome fusions.

One of the complexes involved in telomere maintenance is an enzyme called telomerase, which adds DNA back to telomeres that have become eroded. Several other proteins also regulate this complex. But how the different proteins talk to one another — to keep telomeres the right length, to protect them, and to replicate them during cell division — is poorly understood. Writing on page 1013 of this issue and in Current Biology, respectively, Loayza and de Lange1 and Colgin et al.2 describe a crucial feature of the process by which telomerase can sense, and thus regulate, the length of individual chromosome ends.

Telomeric DNA is composed of G-rich repeats — reiterations of a short DNA sequence that does not code for protein and is high in guanine (G) nucleic-acid bases. It also has a single-stranded stretch that overhangs the end of the double-stranded (duplex) telomeric region. This overhang is the substrate for telomerase, which elongates chromosome ends by adding G-rich repeats. The importance of telomerase is evident from studies of yeast and human cells in which reductions in telomerase levels produce a steady decline in telomere length that eventually blocks cell division. Not surprisingly, then, telomerase is highly active in systems such as the blood and reproductive system, which rely on continuous replenishment through cell proliferation3. Much to the interest of cancer biologists, telomerase levels are also increased in most human tumours, providing a potential target for the development of anticancer drugs4.

In normal cells, telomerase activity is carefully controlled by several mechanisms. For instance, subunits that are part of the telomerase complex itself can positively regulate the enzyme, for example by mediating recruitment of the complex to chromosome ends5. Surprisingly, proteins that bind to the duplex region of the telomere can also be potent regulators, even though they do not appear to associate physically with telomerase. These duplex-binding proteins — which include Rap1 in budding yeast and the TRF1 and TRF2 proteins in human cells — can 'count' the number of G-rich repeats and, when telomeres become overly long, inhibit further telomerase activity6,7.

Missing from this elegant proposal for telomere-length regulation, however, is an explanation for how information from the duplex portion of the telomere is relayed to the very tip of the chromosome — the site of telomerase action. To address this, Loayza and de Lange1 and Colgin et al.2 turned to a recently discovered human protein called POT1 (ref. 8). POT1 is related to well-known proteins that bind single-stranded telomeric DNA in ciliates and budding yeast9,10, and it itself binds to single-stranded telomeric substrates in vitro8. Loayza and de Lange show that, consistent with this biochemical property, POT1 is present at the tips of human chromosomes, and that this association is reduced when the length of the G-rich single-stranded overhang is decreased. But POT1 does not rely solely on its nucleic-acid-binding properties to bring it to telomeres. It is also found in a complex with the duplex-binding protein TRF1 (along with other TRF1-associated proteins, collectively referred to as the TRF1 complex). So, POT1 interacts with two different sites on telomeres: the extreme terminus, and along the duplex region.

What happens if POT1 loses its ability to interact with one of these two sites? Loayza and de Lange tested this by removing the DNA-binding portion of the protein, creating a derivative called POT1(ΔOB). This derivative still associated with the TRF1 complex, and hence with duplex telomeric DNA. But it could not interact with the single-stranded telomeric ends, and this resulted in a profound disruption of telomere length, with telomeres becoming rapidly and extensively elongated. These observations establish that POT1, like TRF1, is a negative regulator of telomere length. More significantly, because POT1 can interact with both the site of telomerase action and the duplex portion of the chromosome, it could be the missing link in the telomere-repeat-counting model.

Loayza and de Lange therefore propose that information about telomere length, which is measured by TRF1 through its ability to bind duplex telomeric repeats, is transferred to the tip of the telomere through the interaction between POT1 and TRF1. This interaction presumably affects the amount of POT1 that is loaded onto single-strand overhangs: the more telomeric repeats there are, the more POT1 is transferred to the telomere tip, and the greater is the reduction in telomerase activity (Fig. 1). The next question is how POT1 relays information to telomerase. It might act as a negative regulator by directly binding to the enzyme complex and inhibiting its catalytic activity. Alternatively, through its DNA-binding activity, POT1 might sequester telomeric DNA and thereby block access of telomerase to its substrate.

Figure 1: Model for telomere length control.

Telomeres are found at the end of linear chromosomes, and their length must be precisely regulated — a process that involves the POT1 protein in human cells1,2. Loayza and de Lange1 have shown that POT1 binds to the TRF1 complex on the double-stranded (duplex) portion of telomeres. TRF1 complexes sense the length of the telomere, and the authors propose that this information is transmitted to telomerase (an enzyme that extends telomeres) via POT1, by transferring POT1 to the single-stranded overhang at the telomere tip. a, When the telomere is long enough, the levels of POT1 on the overhang are high, and telomerase is inhibited. b, When the telomere is too short, little or no POT1 is transferred to the end, and telomerase is no longer inhibited, allowing it to add DNA back to the telomere. Colgin et al.2 have proposed that POT1 may also act as a positive regulator of telomerase when present at the single-stranded terminus. It might do so via a direct interaction with telomerase, in an analogous way to how the yeast Cdc13 protein regulates telomerase5,11.

The picture is also far from complete in other respects. For instance, POT1 might not function solely as a negative regulator: Colgin et al.2 propose that it also positively controls telomerase-mediated telomere elongation. The idea that a protein that binds single-stranded telomeric DNA can serve as both a positive and a negative regulator of telomere length has also been established in studies of the yeast Cdc13 protein5,11. Meanwhile, TRF2, like TRF1, binds duplex telomeric DNA and similarly contributes to the feedback mechanism that monitors telomere length7 — yet it does not interact with POT1 (ref. 1). This implies that TRF2 must have a different partner that transmits its signal to the chromosome terminus.

A separate question about POT1 centres on its postulated role in protecting chromosome ends (as opposed to maintaining telomere length) in human cells. Fission yeast lacking POT1 have a severe defect in end protection8 — will the same be true in humans? These open questions suggest that the intense focus on telomere biology is unlikely to abate anytime soon.


  1. 1

    Loayza, D. & de Lange, T. Nature 423, 1013–1018 (2003).

    CAS  Article  ADS  Google Scholar 

  2. 2

    Colgin, L. M., Baran, K., Baumann, P., Cech, T. R. & Reddel, R. R. Curr. Biol. 13, 942–946 (2003).

    CAS  Article  Google Scholar 

  3. 3

    Lee, H.-W. et al. Nature 392, 569–574 (1998).

    CAS  Article  ADS  Google Scholar 

  4. 4

    Damm, K. et al. EMBO J. 20, 6958–6968 (2001).

    CAS  Article  Google Scholar 

  5. 5

    Evans, S. K. & Lundblad, V. Science 286, 117–120 (1999).

    CAS  Article  Google Scholar 

  6. 6

    Marcand, S., Gilson, E. & Shore, D. Science 275, 986–990 (1997).

    CAS  Article  Google Scholar 

  7. 7

    Smogorzewska, A. et al. Mol. Cell. Biol. 20, 1659–1668 (2000).

    CAS  Article  Google Scholar 

  8. 8

    Baumann, P. & Cech, T. R. Science 297, 1171–1175 (2001).

    Article  ADS  Google Scholar 

  9. 9

    Mitton-Fry, R. M., Anderson, E. M., Hughes, T. R., Lundblad, V. & Wuttke, D. S. Science 296, 145–147 (2002).

    CAS  Article  ADS  Google Scholar 

  10. 10

    Peersen, O. B., Ruggles, J. A. & Schultz, S. C. Nature Struct. Biol. 9, 182–187 (2002).

    CAS  PubMed  Google Scholar 

  11. 11

    Chandra, A., Hughes, T. R., Nugent, C. I. & Lundblad, V. Genes Dev. 15, 404–414 (2001).

    CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Vicki Lundblad.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Lundblad, V. Taking the measure. Nature 423, 926–927 (2003).

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.


Sign up for the Nature Briefing newsletter for a daily update on COVID-19 science.
Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing