The POT1–TPP1 telomere complex is a
telomerase processivity factor
Feng Wang1, Elaine R. Podell2, Arthur J. Zaug2, Yuting Yang1, Paul Baciu1, Thomas R. Cech2& Ming Lei1
Telomeres were originally defined as chromosome caps that prevent the natural ends of linear chromosomes from
undergoing deleterious degradation and fusion events. POT1 (protection of telomeres) protein binds the single-stranded
G-rich DNA overhangs at human chromosome ends and suppresses unwanted DNA repair activities. TPP1 is a previously
the crystal structure of a domain of human TPP1 reveals an oligonucleotide/oligosaccharide-binding fold that is structurally
similar to the b-subunit of the telomere end-binding protein of a ciliated protozoan, suggesting that TPP1 is the missing
b-subunit of human POT1 protein. Telomeric DNA end-binding proteins have generally been found to inhibit rather than
that POT1–TPP1 switches from inhibiting telomerase access to the telomere, as a component of shelterin, to serving as a
processivity factor for telomerase during telomere extension.
Telomeres, the specialized DNA–protein complexes found at the
termini of all linear eukaryotic chromosomes, protect chromosomes
from degradation and end-to-end fusion1. Telomeric DNA typically
beyond its complement to form a 39 overhang. In most eukaryotes,
telomere length is maintained by telomerase, a specialized reverse
transcriptase that adds telomeric DNA to the 39 ends of chromo-
somes to ensure complete genome replication2. Telomerase is
strongly upregulated in most cancer cells and has been studied as a
plausible anti-cancer target3.
A six-protein complex is thought to protect the telomeres of
human chromosomes. TRF1 and TRF2 directly bind double-
stranded telomeric DNA4,5, POT1 directly binds the single-stranded
39 extension at the chromosome end6–11, and these are bridged
12–19). The sixth protein, RAP1, binds mostly to TRF2 (refs 18, 20).
Two functions have been proposed for this complex: protecting the
natural chromosome end from being mistaken for a broken end and
by sequestration of its telomeric DNA substrate. Both functions of
this complex are captured by the name shelterin21.
the telomere, we were surprised to find that it increases both the
activity and processivity of core telomerase. This is the first protein
complex shown to substantially activate telomerase processivity. The
crystal structure of TPP1 shows high structural similarity to the
b-subunit of TEBP (telomere end-binding protein) from Oxytricha
b dimer is more conserved evolutionarily than had been expected.
TPP1 and POT1 form ternary complexes with ssDNA
Recombinant human TPP1 protein with an N-terminal deletion,
TPP1(902544) (Fig. 1a), was overexpressed and purified from
Escherichia coli. TPP1(902544) was chosen because the 87
N-terminal residues of TPP1 are functionally dispensable in human
cells12–14and are not conserved among TPP1 proteins of different
organisms12,13. For simplicity, we hereafter use TPP1 to represent
TPP1(902544) unless stated otherwise.
An 18-nucleotide single-stranded telomeric DNA (primer a,
(TTAGGG)3) was incubated with increasing amounts of TPP1 with
or without POT1, and binding was analysed by electrophoretic
mobility shift assay (EMSA). POT1 protein bound to the DNA,
whereas TPP1 on its own did not, even at a high protein concentra-
tion (375nM) (Fig. 1b, lanes 1–5). When TPP1 was added to the
POT1–DNA mixture, however, an additional complex formed that
migrated above the POT1–DNA complex (Fig. 1b, lanes 6–11). By
two criteria, this more slowly migrating complex contained TPP1.
First, two size variants of TPP1, both of which contain the POT1-
plexes (Fig. 1c). Second, addition of an anti-His antibody confirmed
that the slower complex contained His-tagged TPP1 (Fig. 1d). The
Fig. 1b), suggesting higher affinity.
The equilibrium dissociation constants (Kd) of the protein com-
plexes with various telomeric single-stranded (ss)DNAs were deter-
mined. Primer a (Fig. 2a) bound POT1 with a Kdof 26nM, and the
stability of this complex was increased sixfold by addition of TPP1
(Supplementary Fig. 1). Primers a5 and a3 (Fig. 2a) contain single-
nucleotide substitutions that force POT1 to bind either to a 59 site
(corresponding to an internal site on a long telomeric G-overhang)
or to a 39-proximal site (corresponding to end-capping)26. The
POT1–TPP1 complex showed a substantial preference for 39-end
binding (Kd50.7 versus 7.4nM; compare circles in Fig. 2b, c).
Notably, the dissociation constant of POT1–a5 is also tenfold higher
than that of POT1–a3 (8.3 versus 89nM) (compare triangles in
Fig. 2b, c), suggesting that the 39 end preference of the POT1–
TPP1 complex is mainly dictated by POT1. Measurements of the
1Department of Biological Chemistry, University of Michigan Medical School, MSRBIII 5301D, 1150 W. Medical Center Drive, Ann Arbor, Michigan 48109, USA.2Howard Hughes
Medical Institute, Department of Chemistry and Biochemistry, University of Colorado, Boulder, Colorado 80309, USA.
Vol 445|1 February 2007|doi:10.1038/nature05454
of the POT1–DNA complex was decreased by addition of equimolar
TPP1 (Supplementary Fig. 1). Furthermore, the complexes with a3
were kinetically about tenfold more stable than those with a5
(Supplementary Fig. 1). Taken together, these data indicate two
POT1–TPP1 binding modes on telomeric DNA, a lower affinity
one at internal sequences and a higher affinity one at the 39 end.
Structural conservation between TPP1 and TEBPb
Functional and structural studies have established that POT1 is the
human homologue of the O. nova TEBP a-subunit6,7,10,25. Although
O. nova and a related ciliate, Stylonychia mytilis27, the DNA-binding
properties of POT1–TPP1 closely resembled those of O. nova
TEBPa–b28, consistent with TPP1 being the human homologue of
TEBPb. In addition, their domain organizations revealed clear sim-
ilarities (Fig. 1a). First, both TPP1 and TEBPb use a central region
(PBD in TPP1 and aBD in TEBPb) to interact with the carboxy-
terminal domains of their binding partners (POT1-C and TEBPa-
C). Second, primary sequence analysis of the N-terminal domain of
TPP1 (residues 90–250) predicted a secondary structure pattern of
a-b-b-b-a-b-b-a (data not shown), where the bold region is char-
acteristic of oligonucleotide/oligosaccharide-binding (OB) folds
found in many telomere-binding proteins including TEBPb10,23.
Third, the C-terminal domains of both TPP1 and TEBPb (TPP1-C
and TEBPb-C) are not involved in the interaction with the
a-subunits (POT1 and TEBPa) and have evolved to have distinct
For crystallization studies, we first purified the N-terminal frag-
ment TPP1-N (Fig. 1a), which can interact with POT1 and would
correspond to the TEBPb fragment in the TEBPa–b–ssDNA crystal
structure23. However, TPP1-N was unstable by itself. Limited pro-
teolysis and matrix-assisted laser desorption/ionization (MALDI)
mass spectrometry identified a protease-resistant core domain of
TPP1-N containing residues 90–260 (Supplementary Fig. 2); this
domain corresponds to the predicted OB fold (Fig. 1a).
1 2 3 4 5 6 7 8 9 1011
1 2 3 4 5 6 7 8 9 1011
– – + – – +
– + – – + –
– – – + + +
2 3 4 5 6
ssDNA-binding domain (POT1-N)
ssDNA-binding domain (TEBPα-N)
Figure 1 | TPP1 binds to the POT1–ssDNA complex and enhances the
POT1–ssDNA interaction. a, Human POT1–TPP1 and O. nova TEBPa–b
complexes share similar domain organization. In POT1 and TEBPa, the
TPP1/TEBPb-binding domains are in light blue. In TPP1 and TEBPb, the
N-terminal OB folds are in orange, the central a-subunit-binding region
(PBD in TPP1 and aBD in TEBPb) is in cyan, and the C-terminal domains
are in yellow. Numbers indicate amino acid positions at the boundaries of
various subdivisions. b,TPP1 requiresPOT1 inorderto interact stablywith
TPP1 (2, 10 and 50nM in lanes 2–4, 2–100 nM in lanes 6–11) in the absence
or presence of 150nM POT1. c, Both TPP1 and TPP1-N generate super-
shifted species but with different mobility; use of a longer primer enhanced
the separation. d, The addition of anti-His antibody TPP1 detected the
existence of His-tagged TPP1 in the slower migrating complex.
05 101520 2530
150 200 250
Fraction bound (%)
Fraction bound (%)
Primer a3Primer a5
Figure 2 | The POT1–TPP1 complex binds to the single-stranded telomeric
overhang with 39 end preference. a, Sequences of primers a, a5 and a3. The
boldletters arethepoint mutations. The POT1-binding sitesare denotedby
boxes26. b, c, Equilibrium binding curves for primers a3 (b) and a5
(c) binding to POT1 and the POT1–TPP1 complex. The solid and dashed
lines represent theoretical binding curves fit to the data for POT1 and
POT1–TPP1, respectively.The calculated equilibrium dissociationconstant
(Kd) values are indicated. The binding curves and Kdvalues for primers a, b
and hT10 (TTAGGGTTAG) are shown in Supplementary Fig. 1.
NATURE|Vol 445|1 February 2007
Recombinant TPP1(90–250) expressed from E. coli was crystallized,
and the structure was solved by single anomalous dispersion (SAD)
and refined to a resolution of 2.7A˚(Supplementary Table 1). The
final model contains residues 90–243 (Fig. 3a).
The structure of TPP1(90–250) reveals a typical OB-fold architec-
ture comprising a highly curved five-stranded b-barrel (Fig. 3a)31,32.
Hereafter, we will refer to TPP1(90–250) as TPP1-OB (Fig. 1a). An
unbiased search for structurally homologous proteins using Dali33
revealed that the structure of TPP1-OB is most similar to that of
the OB fold of the O. nova TEBP b-subunit23. The two structures
can be superimposed with a root-mean-square deviation of 2.0A˚in
the positions of 144 equivalent Ca atoms (Fig. 3b). Notably, this
structurally conserved region includes not only the central b-barrel,
but also three peripheral a-helices, suggesting that TPP1 and TEBPb
are homologous proteins (Fig. 3b); other OB folds, such as that of
structural similarity, the OB folds of TPP1 and TEBPb share several
theOB folds ofPOT1andTEBPa,adopts anextended conformation
and packs across one side of the b-barrel, forcing helix aC to cap the
bottom of the barrel (Fig. 3b). Second, helix aB is in a modified
position, rotated almost 90u relative to the orientation normally
observed in OB folds. Taken together, these structural similarities
strongly support the notion that TPP1 is the human homologue of
O. nova TEBPb. Given that TPP1 has been identified in many other
eukaryotes12,13, TPP1/TEBPb may be an evolutionarily conserved
Despite the high degree of structural conservation, the sequences
of OB folds of TPP1 and TEBPb are markedly divergent and share
only 11% identity (Supplementary Fig. 3). Significant sequence and
structural variation is particularly evident in the connecting loop
regions. TPP1-OB has a very long loop (20 residues), L12, between
barrel (Fig. 3a, b). In contrast, strands b1 and b2 of TEBPb are
connected by a short two-residue turn (Fig. 3b). These marked var-
iances in the loop regions explain the failure to detect the similarity
between these OB folds by bioinformatics.
POT1–TPP1 is a telomerase processivity factor
Weinvestigated the ability oftelomerase to extend thePOT1–TPP1–
ssDNA ternary complex, expecting some inhibition consistent with
the shelterin model21. Human core telomerase was reconstituted in
vitro and immunopurified via the haemagglutinin (HA) tag on the
telomerase catalytic subunit (TERT). Primer a5 has a single-nucleo-
tide mutation that forces POT1tobind toits 59end (Fig.2a), leaving
a telomerase-extendible 39 tail26. Addition of POT1 and TPP1 to
primer a5 markedly increased the telomerase product size distri-
bution. Primer a5 was extended via more than 30 cycles of template
copying (Fig. 4a, lane 4), whereas in the absence of the POT1–TPP1
complex, the first three cycles accounted for most of the extension
(lane 1). Under conditions of vast primer excess, longer extension
products result from processive extension, not from rebinding of
previously extended products34. We confirmed that this condition
still pertained in the case of the ternary complexes by showing that
the extension was independent of concentration over a 2,000-fold
range (Supplementary Fig. 4).
These data emphasize the longer extension products, because32P-
GMP incorporation increases with product size. Quantification
showed that POT1–TPP1 provided a threefold increase in activity
(Fig. 4b) and, after dividing each product by the amount of GMP
incorporated, a fourfold increase in processivity (Fig. 4c). R1/2, the
number of repeats synthesized before half of the chains have disso-
with the POT1–TPP1–DNA complex. Because this fourfold increase
in processivity is cumulative, it has a very large effect on the produc-
tion of longer products (see double-headed arrows in Fig. 4c–e).
POT1 by itself produced a more modest stimulation of processivity,
small molecule accompanying the protein. However, when we frac-
tionated TPP1 by gel filtration chromatography, the processivity
activity clearly co-migrated with the main protein peak (Sup-
plementary Fig. 5), indicating that the enhanced processivity was
does primer b ((GGTTAG)3)26. As expected, the addition of POT1
almost completely blocked the extension of these primers (Fig. 4a,
lanes 6 and 10). When TPP1 was also added, however, telomere
extension was rescued and processivity was increased five- to sixfold
(Fig. 4a, lanes 8 and 12; see also Fig. 4d, e). Totest the possibility that
TPP1 induced sliding of POT1 from the preferred 39 binding site to
the 59 site, we mapped the position of the leading edge of the POT1–
TPP1 complex on the DNA using snake-venom phosphodiesterase
I, which degrades ssDNA exonucleolytically from the 39 end.
Complexes of primer a3 or primer a with the POT1–TPP1 hetero-
a local change in accessibility relative to the POT1–DNA complexes
where the 39 end of the DNA protrudes from the protein (Sup-
plementary Fig. 6). Although TPP1 did not relocate the bulk of
POT1 to an internal site on the DNA, we propose that the proteins
are in fact binding internally in a minority of the complexes26; the
resulting 39 overhang is then extended by telomerase with the
enhanced processivity characteristic of POT1–TPP1 complexes.
Figure 3 | The crystal structure of TPP1-OB indicates that TPP1 is the
homologue of O. nova TEBPb. a, Ribbon diagram of TPP1-OB with
structure elements are labelled. b, Superposition of TPP1-OB on the crystal
structure of the OB fold of TEBPb23. TPP1 is in red and TEBPb in blue.
c, Superposition of TPP1-OB on the crystal structure of the first OB fold of
the program Pymol (http://pymol.sourceforge.net).
NATURE|Vol 445|1 February 2007
therefore foldsintoHoogsteen base-pairedG-quadruplexstructures,
which are inhibitory for telomerase extension (Fig. 4a, lane 13)35.
POT1 can trap an open form of this DNA, allowing telomerase
increase in telomerase activity seen with all the primers (Fig. 4b),
but telomerase still stalled after every nucleotide added to the
G-quadruplex. POT1–TPP1 again stimulated highly processive
extension by telomerase (lane 16).
We next asked whether the enhanced telomerase processivity and
activity were dependent on the POT1–TPP1 interaction by using a
panel of POT1 and TPP1 deletion mutant proteins. POT1-N, which
lacks the TPP1 interaction domain, and TPP1-OB, which lacks the
POT1 interaction domain, both failed to endow telomerase with
increased processivity and activity (Supplementary Fig. 7). Only
when both proteins had intact interaction domains was the proces-
sivity greatly stimulated (Supplementary Fig. 7), confirming the
important role of the POT1–TPP1 interaction in this activity. The
purified POT1 interaction domain of TPP1, TPP1-PBD, was insuf-
ficient to activate telomerase in the presence of POT1 (Supplemen-
tary Fig. 8).
E. coli DNA polymerase achieves high processivity by means of an
accessory protein that serves as a ‘sliding clamp’, encircling the DNA
and preventing dissociation36. By analogy, the POT1–TPP1 complex
might move with telomerase, binding the DNA just upstream from
its 39 end and inhibiting dissociation. Given the DNA sequence-
specificity of POT1–TPP1 binding, the protein would not slide con-
tinuously along the DNA but would ratchet in 6- or 12-nucleotide
steps. On the other hand, ssDNA is intrinsically much more flexible
than double-stranded (ds)DNA. Thus, a clamp would not need to
slide or ratchet to keep the ssDNA associated with telomerase, but
could instead remain fixed while the newly synthesized telomeric
for future research is whether a single POT1–TPP1 complex clamp is
sufficient for increased processivity, or whether the proteins must
coat the elongating telomeric DNA.
In normal human somatic cells that lack telomerase, telomeres
shrink by about 30–100base pairs per replication cycle37. This pro-
vides a plausible estimate for the amount of DNA synthesized by
telomerase at each chromosome end, and is similar to more direct
measurements in yeast38. Whether such extension is processive or
distributive in vivo is unknown. However, we note that the telomer-
ase processivity achieved here in the presence of POT1–TPP1 is
around four repeats or 24nucleotides, which would mean that one
or a few rounds of telomerase extension per cell cycle would be
sufficient to maintain human telomeres.
– – + +
– + – +
– – + +
– + – +
– – + +
– + – +
– – + +
– + – +
a5 a3b AGGG-a
911 13 15 17 19 21
1357911 13 15 17 19 21
1357911 13 15 17 19 21
Primer a5 processivity
Log10 [normalized activity]
Log10 [normalized activity]
Log10 [normalized activity]
Primer a3 processivity
Primer b processivity
Figure 4 | The POT1–TPP1 complex functions as a telomerase processivity
factor. a, Direct telomerase activity assays34were performed with 100nM
primer a5 (lanes 1–4), a3 (lanes 5–8), b (lanes 9–12), or AGGG-a (lanes
13–16) in the presence of a saturating concentration of POT1, TPP1, or
POT1–TPP1. Reaction products were then analysed by gel electrophoresis
results. b, Quantification of total DNA synthesis relative to synthesis in the
presence of protein buffer alone. c–e, Activity in each repeat shown in a was
measured, corrected for the number of radiolabelled nucleotides
k is the slope andR1/2is thenumber ofrepeats synthesized before halfof the
chains have dissociated, analogous to t1/2in radioactive decay.
NATURE|Vol 445|1 February 2007
POT1–TPP1 complex can also function as a positive telomerase pro- Download full-text
cessivity factor. We propose a three-state model of telomere length
regulation that can reconcile the two apparently opposite functions
ofPOT1–TPP1. (1)When POT1–TPP1covers the39terminus ofthe
G-overhang, it sequesters the telomere and prevents binding of tel-
omerase. (2) POT1–TPP1 is removed from its high-affinity 39 bind-
ing site by an unidentified mechanism, which might, for example,
involve post-translational modification and disruption of the shel-
terin complex. (3) The POT1–TPP1 complex then serves as a telo-
merase processivity factor during telomere extension. As the
telomere is elongated and reaches a certain threshold, the newly
synthesized repeats bind shelterin complexes, the 39 end of the over-
hang is re-bound by POT1–TPP1, and further telomerase extension
is inhibited (back to state 1). Further work will be needed to under-
stand how switching between such telomere and telomerase com-
plexes is achieved and regulated in vivo.
Details of recombinant protein cloning, expression and purification can be
assay were purchased from IDT and Invitrogen and 59-end-labelled using poly-
nucleotide kinase. Protein binding proceeded for 1h at room temperature in
90mM NaCl, 50mM Tris-HCl, pH8.0, 6mM dithiothreitol and 5% glycerol,
followed by EMSA on native 6% polyacrylamide gels10. For in vitro telomerase
activity assays26, C-terminal HA-tagged human TERT was expressed from
phTERT-HA2 and hTR from phTR34using the TnT quick-coupled transcrip-
tion/translation system (Promega), and the reconstituted core telomerase was
affinity-purified on anti-HA F7-agarose beads (Santa Cruz Biotechnology)26.
Details of crystallization, data collection and structure determination of TPP1-
OB can be found in Supplementary Methods.
Received 17 August; accepted 17 November 2006.
Published online 21 January 2007.
Shay, J. W. & Wright, W. E. Telomerase: a target for cancer therapeutics. Cancer
Cell 2, 257–265 (2002).
Chong, L. et al. A human telomeric protein. Science 270, 1663–1667 (1995).
Broccoli, D., Smogorzewska, A., Chong, L. & de Lange, T. Human telomeres
contain two distinct Myb-related proteins, TRF1 and TRF2. Nature Genet. 17,
Baumann, P. & Cech, T. R. Pot1, the putative telomere end-binding protein in
fission yeast and humans. Science 292, 1171–1175 (2001).
Baumann, P., Podell, E. & Cech, T. R. Human Pot1 (protection of telomeres)
protein: cytolocalization, gene structure, and alternative splicing. Mol. Cell. Biol.
22, 8079–8087 (2002).
Hockemeyer, D., Daniels, J. P., Takai, H. & de Lange, T. Recent expansion of the
telomeric complex in rodents: Two distinct POT1 proteins protect mouse
telomeres. Cell 126, 63–77 (2006).
Wu, L. et al. Pot1 deficiency initiates DNA damage checkpoint activation and
aberrant homologous recombination at telomeres. Cell 126, 49–62 (2006).
10. Lei, M., Podell, E. R. & Cech, T. R. Structure of human POT1 bound to telomeric
single-stranded DNA provides a model for chromosome end-protection. Nature
Struct. Mol. Biol. 11, 1223–1229 (2004).
11.Loayza, D. & de Lange, T. POT1 as a terminal transducer of TRF1 telomere length
control. Nature 423, 1013–1018 (2003).
12. Ye, J. Z. et al. POT1-interacting protein PIP1: a telomere length regulator that
recruits POT1 to the TIN2/TRF1 complex. Genes Dev. 18, 1649–1654 (2004).
13. Liu, D. et al. PTOP interacts with POT1 and regulates its localization to telomeres.
Nature Cell Biol. 6, 673–680 (2004).
14. Houghtaling, B. R., Cuttonaro, L., Chang, W. & Smith, S. A dynamic molecular link
between the telomere length regulator TRF1 and the chromosome end protector
TRF2. Curr. Biol. 14, 1621–1631 (2004).
15. Kim, S. H., Kaminker, P. & Campisi, J. TIN2, a new regulator of telomere length in
human cells. Nature Genet. 23, 405–412 (1999).
16. Liu, D. O’connor, M. S., Qin, J. & Songyang, Z. Telosome, a mammalian telomere-
associated complex formed by multiple telomeric proteins. J. Biol. Chem. 279,
17. Ye, J. Z. et al. TIN2 binds TRF1 and TRF2 simultaneously and stabilizes the TRF2
complex on telomeres. J. Biol. Chem. 279, 47264–47271 (2004).
18. O’Connor, M. S., Safari, A., Xin, H., Liu, D. & Songyang, Z. A critical role for TPP1
and TIN2 interaction in high-order telomeric complex assembly. Proc. Natl Acad.
Sci. USA 103, 11874–11879 (2006).
19. Kim, S. H. et al. TIN2 mediates functions of TRF2 at human telomeres. J. Biol.
Chem. 279, 43799–43804 (2004).
20. Li, B., Oestreich, S. & de Lange, T. Identification of human Rap1: implications for
telomere evolution. Cell 101, 471–483 (2000).
21. de Lange, T. Shelterin: the protein complex that shapes and safeguards human
telomeres. Genes Dev. 19, 2100–2110 (2005).
22. Hicke, B. J., Celander, D. W., MacDonald, G. H., Price, C. M. & Cech, T. R. Two
versions of the gene encoding the 41-kilodalton subunit of the telomere binding
protein of Oxytricha nova. Proc. Natl Acad. Sci. USA 87, 1481–1485 (1990).
23. Horvath, M. P., Schweiker, V. L., Bevilacqua, J. M., Ruggles, J. A. & Schultz, S. C.
Crystal structure of the Oxytricha nova telomere end binding protein complexed
with single strand DNA. Cell 95, 963–974 (1998).
24. Gray, J. T., Celander, D. W., Price, C. M. & Cech, T. R. Cloning and expression of
genes for the Oxytricha telomere-binding protein: specific subunit interactions in
the telomeric complex. Cell 67, 807–814 (1991).
of human POT1: a nonamer 59-TAGGGTTAG-39 minimal binding site, sequence
specificity, and internal binding to multimeric sites. J. Biol. Chem. 279,
26. Lei, M., Zaug, A. J., Podell, E. R. & Cech, T. R. Switching human telomerase on and
off with hPOT1 protein in vitro. J. Biol. Chem. 280, 20449–20456 (2005).
27. Fang, G. W. & Cech, T. R. Molecular cloning of telomere-binding protein genes
from Stylonychia mytilis. Nucleic Acids Res. 19, 5515–5518 (1991).
28. Fang, G., Gray, J. T. & Cech, T. R. Oxytricha telomere-binding protein: separable
DNA-binding and dimerization domains of the a-subunit. Genes Dev. 7, 870–882
29. Fang, G. & Cech, T. R. The b subunit of Oxytricha telomere-binding protein
promotes G-quartet formation by telomeric DNA. Cell 74, 875–885 (1993).
30. Paeschke, K., Simonsson, T., Postberg, J., Rhodes, D. & Lipps, H. J. Telomere end-
binding proteins control the formation of G-quadruplex DNA structures in vivo.
Nature Struct. Mol. Biol. 12, 847–854 (2005).
31. Dietmann, S. & Holm, L. Identification of homology in protein structure
classification. Nature Struct. Biol. 8, 953–957 (2001).
32. Theobald,D.L.,Mitton-Fry, R.M.&Wuttke,D.S.Nucleicacidrecognition byOB-
fold proteins. Annu. Rev. Biophys. Biomol. Struct. 32, 115–133 (2003).
33. Holm, L. & Sander, C. Database algorithm for generating protein backbone and
side-chain co-ordinates from a Ca trace application to model building and
detection of co-ordinate errors. J. Mol. Biol. 218, 183–194 (1991).
34. Chen, J. L. & Greider, C. W. Determinants in mammalian telomerase RNA that
mediate enzyme processivity and cross-species incompatibility. EMBO J. 22,
35. Zaug, A. J., Podell, E. R. & Cech, T. R. Human POT1 disrupts telomeric
G-quadruplexes allowing telomerase extension in vitro. Proc. Natl Acad. Sci. USA
102, 10864–10869 (2005).
36. Kuriyan, J. & O’Donnell, M. Sliding clamps of DNA polymerases. J. Mol. Biol. 234,
is arrested in immortal cells which express telomerase activity. EMBO J. 11,
is achieved via a switch between telomerase-extendible and -nonextendible
states. Cell 117, 323–335 (2004).
Supplementary Information is linked to the online version of the paper at
project; N. F. Lue for the His-SUMO protein expression vector; Z. Songyang and
T. de Lange for TPP1 cDNA; D. Yoder of beamline 23-ID at APS for assistance with
data collection; and J. L. Chen and C. W. Greider for the human TERT and TER
plasmids. Work in the laboratory of M.L. is supported by the American Cancer
Society and the Sidney Kimmel Foundation. E.R.P., A.J.Z. and T.R.C. are supported
by the Howard Hughes Medical Institute.
Author Contributions F.W. is responsible for the bulk of the experiments; Y.Y. for
structural determination of TPP1-OB; P.B. for crystallization of TPP1-OB; E.R.P. and
A.J.Z. for the telomerase activity assays and some of the EMSA experiments; and
T.R.C. and M.L. contributed to overall design and interpretation of the studies.
been deposited in the RCSB Protein Data Bank with accession code 2I46. Reprints
declare no competing financial interests. Correspondence and requests for
materials should be addressed to M.L. (email@example.com).
NATURE|Vol 445|1 February 2007