Molecular Cell
Article
A RAP1/TRF2 Complex Inhibits Nonhomologous
End-Joining at Human Telomeric DNA Ends
Nancy S. Bae1 and Peter Baumann1,2,*
1
Stowers Institute for Medical Research, Kansas City, MO 64110, USA
Department of Biochemistry and Molecular Biology, University of Kansas Medical Center, Kansas City, KS 66160, USA
*Correspondence: [email protected]
DOI 10.1016/j.molcel.2007.03.023
2
SUMMARY
The mechanisms by which telomeres are distinguished from DNA double-strand breaks are
poorly understood. Here we have defined the
minimal requirements for the protection of telomeric DNA ends from nonhomologous endjoining (NHEJ). Neither long, single-stranded
overhangs nor t loop formation is essential to
prevent NHEJ-mediated ligation of telomeric
ends in vitro. Instead, a tandem array of 12 telomeric repeats is sufficient to impede illegitimate
repair in a highly directional manner at nearby
DNA ends. The polarity of end protection is
consistent with the orientation of naturally occurring telomeres and is well suited to minimize
interference between chromosome capping
and the repair of DNA double-strand breaks in
subtelomeric sequences. Biochemical fractionation and reconstitution revealed that telomere
protection is mediated by a RAP1/TRF2 complex, providing evidence for a direct role for
human RAP1 in the protection of telomeric DNA
from NHEJ.
INTRODUCTION
The efficient repair of DNA double-strand breaks (DSBs) is
essential for the maintenance of genome stability in all
organisms. Homologous recombination (HR) and nonhomologous end-joining (NHEJ) are the two principal pathways that repair such lesions (Wyman and Kanaar, 2006).
Homologous recombination is a highly accurate repair
pathway that employs a homologous template sequence
to restore information lost at the site of a break (West,
2003). It is predominantly active during S and G2 phases
of the cell cycle when a sister chromatid is available as a
template. In NHEJ, the two DNA ends are simply rejoined,
a mechanism that is error prone as the damaging event or
subsequent processing may result in the loss of nucleotides on either side of the break (Burma et al., 2006).
NHEJ is active throughout the cell cycle but is thought to
be the dominating pathway in G1 when no sister chromatid is available as template for homologous recombination
(Ferreira and Cooper, 2004).
An interesting problem arises in cells with linear genomes, as the natural ends of chromosomes must be
distinguished from DNA breaks and protected from undergoing fusion with each other or sites of DNA damage. Specialized nucleoprotein structures, known as telomeres,
protect chromosome ends from illegitimate repair as well
as from gradual attrition due to the end replication problem
(Cech, 2004; Hug and Lingner, 2006). In most eukaryotes,
telomeres are comprised of GT-rich repeat sequences
with the GT-rich strand always oriented in the 50 to 30 direction toward the chromosome end. This uniform orientation
of a tandem repeat sequence prevents chromosome fusions via homologous recombination. Instead, HR events
can elongate and shorten telomeres and may thereby contribute substantially to telomere maintenance (Lundblad,
2002). In contrast, NHEJ between telomeres results in genomic instability, mitotic catastrophe, and frequently cell
death (Karlseder et al., 1999; van Steensel et al., 1998).
The mechanism by which NHEJ is prevented from acting
on chromosome ends has been a matter of considerable
interest for some time. Over the past 20 years, proteins
that specifically localize to telomeres have been identified
in a number of organisms. In vertebrates, double-stranded
telomeric repeats are directly bound by the homeodomain
proteins TRF1 and TRF2 (Bilaud et al., 1997; Broccoli et al.,
1997; Court et al., 2005; Chong et al., 1995). These recruit
TIN2 (Kim et al., 1999), which via TPP1 (Houghtaling et al.,
2004; Liu et al., 2004; Ye et al., 2004) can form a bridge with
POT1 (Baumann and Cech, 2001), the protein that specifically binds to the single-stranded 30 overhang at the end of
the G-rich strand (Baumann et al., 2002; Loayza and de
Lange, 2003). TRF2 also interacts directly with RAP1 and
POT1 (Li et al., 2000; Yang et al., 2005). These six proteins
have been referred to as the shelterin complex (de Lange,
2005), indicating a role in protecting the chromosome end
from repair activities, as much as in negatively regulating
telomerase by sequestering its DNA substrate in a closed
conformation.
Surprisingly, numerous proteins with known roles in
DNA repair are also involved in telomere maintenance.
The Ku heterodimer, the Mre11-Rad50-Nbs1 (MRN) complex, ATM, and several other proteins described as damage sensors, checkpoint proteins, and repair factors have
been implicated in normal telomere function (d’Adda di
Fagagna et al., 2004). Further blurring the distinction
between telomere maintenance and DNA repair are the
recent findings that the telomere-binding protein TRF2
Molecular Cell 26, 323–334, May 11, 2007 ª2007 Elsevier Inc. 323
Molecular Cell
Inhibition of NHEJ by Human RAP1/TRF2 Complex
and its fission yeast homolog Taz1 may be directly involved in DSB repair in addition to their roles at telomeres
(Bradshaw et al., 2005; Miller and Cooper, 2003).
Over the past two decades several mechanisms for
chromosome end protection have been proposed. Possibly the most elegant solution to establishing an inaccessible state of a telomere is the remodeling of linear DNA into
telomeric loops (t loops), structures in which the terminal
single strand invades internal regions of the same telomere. Such structures have been isolated from several
organisms and visualized by electron microscopy (de
Lange, 2004; Griffith et al., 1999). Purified TRF2 can facilitate t loop formation in vitro (Stansel et al., 2001) and may
do so alone or as part of a complex in vivo. It is presently
unclear whether all telomeres or only a subset adopt the
t loop conformation, and whether these structures persist
for most of the cell cycle or appear transiently, possibly as
intermediates of telomere rapid deletion events (Ancelin
et al., 2002; Lustig, 2003; Wang et al., 2004). Chromatin
immunoprecipitation experiments revealed that DNA
damage response factors are recruited to telomeres in
G2 (Verdun et al., 2005; Verdun and Karlseder, 2006).
These results support a model in which chromosome
ends are transiently recognized as DNA breaks after replication and are then converted into a stable structure by
a process that requires telomere-binding proteins and
the homologous recombination machinery.
Expression of mutant proteins and RNAi-mediated
knockdown experiments have implicated several proteins
in telomere capping. For example, knockdown of human
POT1 was found to cause an increase in chromosome end
fusions (Hockemeyer et al., 2005; Veldman et al., 2004;
Yang et al., 2005). Similarly, knockout of POT1a and
POT1b in mouse embryonic fibroblasts resulted in chromosomal abnormalities (He et al., 2006; Hockemeyer
et al., 2006; Wu et al., 2006). Surprisingly, a comparable
increase in chromosome end fusions is also observed in
cells defective in Ku, DNA-PKcs, or Artemis, all proteins
involved in DSB repair by NHEJ (reviewed in Ferreira
et al. [2004]). The mechanism by which these proteins
function in telomere capping has remained largely enigmatic, but a high-affinity interaction between Ku and TRF1
has been suggested to prevent NHEJ uniquely at telomeres (Hsu et al., 2000). Consistent with such a mechanism, telomere fusions have been observed following
conditional deletion of TRF1 in ES cells (Iwano et al.,
2003), although another study found no evidence of telomere uncapping in chromosome spreads from TRF1/
blastocysts (Karlseder et al., 2003).
Arguably, the most dramatic incidence in chromosome
fusions has been associated with expression of TRF2DBDM,
a dominant-negative mutant of TRF2 (van Steensel et al.,
1998). Displacement of endogenous TRF2 from telomeres
by TRF2DBDM resulted in abundant ligase IV-dependent
telomere fusions (Smogorzewska et al., 2002). A conditional knockout of TRF2 in mouse embryonic fibroblasts
revealed an even more dramatic phenotype with virtually
all telomeres engaged in end-to-end fusions (Celli and
de Lange, 2005). These observations argue for a key role
for TRF2 in distinguishing telomeres from DNA DSBs. It is
presently unknown whether the protection of telomeres
from NHEJ by TRF2 is linked to its ability to promote
t loop formation. Single telomere length analysis (STELA)
suggested that a fraction of human telomeres are too short
for t loop formation but may still be protected from undergoing fusions (Baird et al., 2003). These ultrashort telomeres can be comprised of fewer than ten repeats and
are thought to arise as a result of catastrophic telomere
deletion events.
Here we show that fewer than 10 telomeric repeats can
indeed be sufficient to provide protection against illegitimate fusions by NHEJ. Protection is mediated by a
RAP1/TRF2 complex that inhibits DNA-PK- and ligase
IV-mediated end-joining specifically on the 30 side of telomeric repeats. As RAP1 has been shown to function in
NHEJ suppression in fission and budding yeast, our data
indicate that an evolutionarily conserved mechanism may
mediate the protection of telomeres from yeast to man.
RESULTS
Inhibition of NHEJ at Telomeric DNA Ends
An in vitro assay for DSB repair has been developed and
recapitulates the in vivo requirement for NHEJ in that
end-joining is dependent on the Ku heterodimer, DNAPKcs, and XRCC4/ligase IV (Baumann and West, 1998).
This system has been used extensively to study the mechanism of NHEJ and to demonstrate the involvement of
polynucleotide kinase, nucleases, and polymerases in
the processing of ends prior to ligation (Budman and
Chu, 2005; Chappell et al., 2002; Hanakahi et al., 2000;
Huang and Dynan, 2002). To define the minimal requirements for protection of chromosome ends from NHEJ,
we added short arrays of telomeric repeats to a 3 kb nontelomeric DNA substrate (Figure 1A). As a control, arrays
of scrambled telomeric repeats (TGAGTG) were introduced into the same plasmid. When introduced into the
end-joining assay, 32% of the control substrate was
ligated into dimers, trimers, and longer concatemers (Figure 1B, lanes 2–4). This efficiency of end-joining is similar
to what had previously been reported with other DNA substrates (Baumann and West, 1998; Chappell et al., 2002),
suggesting that NHEJ is not inhibited by G-rich repeat
sequences in general. In contrast, only 8% of a substrate
containing 12 telomeric repeats at one end underwent
end-joining, and only dimeric products were observed
(lanes 6–8). It thus appears that the presence of a short
array of telomeric repeats can inhibit NHEJ.
To verify that end-joining in these reactions is indeed
catalyzed by the same proteins that mediate NHEJ in vivo,
a number of control reactions were carried out. Consistent
with a requirement for DNA-PK activity, wortmannin inhibited the end-joining reaction in a concentration-dependent manner (Figure 1C). End-joining with telomeric and
nontelomeric DNA substrates was also reduced following
the addition of blocking antibodies against the Ku70/80
324 Molecular Cell 26, 323–334, May 11, 2007 ª2007 Elsevier Inc.
Molecular Cell
Inhibition of NHEJ by Human RAP1/TRF2 Complex
Figure 1. Protection of Telomeric DNA
Ends from NHEJ
(A) A plasmid containing a tract of telomeric
repeats or scrambled telomeric repeats was
subjected to restriction digest placing the repeat sequence at one end. The end adjacent
to the repeats is referred to as head (H), and
the other end is referred to as tail (T).
(B) Efficient DNA end-joining is catalyzed by extract from human GM00558 cells with scrambled, but not with telomeric, DNA substrates.
Reactions contained EcoRI-cut and 50 -32P
end-labeled DNA with 12 telomeric or scrambled repeats. Inhibition of NHEJ in standard reactions by the DNA-PK inhibitor wortmannin
(C) or by the addition of blocking antibody
raised against Ku (200 ng) (D) or XRCC4
(1.6 ng, lane 1; 2 ng, lane 2; 4 ng, lane 3; 8 ng,
lane 4) (E) is shown.
heterodimer (Figure 1D). A requirement for XRCC4/ligase
IV was confirmed by the addition of blocking antibodies
against XRCC4 as well as by immunodepletion (Figure 1E
and data not shown). We have thus confirmed that endjoining in these reactions is dependent on DNA-PKcs,
Ku, and XRCC4/ligase IV as previously reported (Baumann and West, 1998).
The observation that end-joining was greatly reduced in
the presence of 12 telomeric repeats indicated that activities in the cell extract recapitulate the protection of chromosome ends from NHEJ. Only one end of the DNA substrate used here was telomeric (subsequently referred to
as head); the other was nontelomeric (tail). Ligation products may result from tail-to-tail, head-to-tail, and headto-head fusions (Figure 2A). The orientation of monomers
in the ligation products is revealed by asymmetric digestion with ScaI (site indicated by an S in Figure 2A). As expected, all three types of fusions were recovered with the
control substrate containing scrambled telomeric repeats
(Figure 2B, lane 3). However, only tail-to-tail fusions were
observed with the telomeric substrate (lane 4), demonstrating that telomeric ends did not engage in fusions with
each other (head-to-head) or with nontelomeric ends
(head-to-tail). The absence of trimers and longer conca-
temers is explained by tail-to-tail fusions giving rise to
linear dimers with two telomeric ends, which are thus protected at both ends from further ligation reactions. Indeed,
when a DNA substrate containing telomeric repeats at
both ends was used, end-joining was inhibited (see Figure S1A in the Supplemental Data available with this article
online, lanes 2–4). Our results thus suggest that the presence of a few telomeric repeats is sufficient to prevent
NHEJ at a nearby DNA end.
Minimal Requirements for Inhibition of NHEJ
We next examined how many repeats are required to protect a DNA terminus from NHEJ. For this purpose a series
of linear DNA substrates with 2–24 telomeric repeats at
one end were generated and tested in the end-joining assay. The presence of up to six telomeric repeats did not
impair end-joining as indicated by the formation of dimers,
trimers, and tetramers (Figure 2C, lanes 2, 4, and 6). In
contrast, 12 or more telomeric repeats prevented fusions
involving telomeric ends and only tail-to-tail products
were detected (Figure 2C, lanes 8 and 10). Intermediate
levels of protection were observed in the presence of eight
and ten telomeric repeats (Figure S1B). Varying the
Molecular Cell 26, 323–334, May 11, 2007 ª2007 Elsevier Inc. 325
Molecular Cell
Inhibition of NHEJ by Human RAP1/TRF2 Complex
Figure 2. DNA Substrate Requirements for Inhibition of NHEJ
(A) Schematic of monomeric substrate and dimeric products. Dimers are in tail-to-tail (T-T), head-to-tail (H-T), or head-to-head (H-H) orientation. The
rectangle indicates the position of the repeat sequence; fusion points are marked with black arrows above dimers. S denotes the position of the ScaI
sites used in diagnostic digests to visualize the relative abundance of the three fusion products.
(B) Standard end-joining reactions containing 50 radiolabeled pNB146 and control reactions were phenol/chloroform extracted, treated with ScaI, and
analyzed by agarose gel electrophoresis. Fusion fragments are labeled as in (A); terminal fragments are labeled with open triangles. The filled triangle
indicates the position of a small amount of undigested monomer in lanes 1 and 2.
(C) DNA substrates terminating on the 30 side with the indicated number of TTAGGG repeats were subject to standard end-joining reactions.
(D) Ligation of telomeric DNA ends by T4 DNA ligase. EcoRI-digested pNB146 was incubated in standard end-joining reactions prior to incubation
with T4 DNA ligase (10 units) for 1 hr at 37 C. DNA substrate (linear) and ligation products are labeled on the right; abbreviations are rc (relaxed circle)
and sc (supercoiled circle).
(E) Positioning of telomeric repeats. TTAGGG or scrambled repeats were either positioned at the 30 terminus (lanes 1 and 5) or internally so that the
indicated number of nontelomeric base pairs were present between the last repeat and the terminus.
(F) Schematic of DNA substrates with repeats at 50 or 30 end.
(G) DNA substrates shown in (F) and scrambled controls were incubated in standard end-joining reactions.
number of scrambled telomeric repeats had no effect on
end-joining efficiency (data not shown).
We considered several mechanisms that could prevent
NHEJ of telomeric ends. First, the machinery responsible
for NHEJ may not assemble on telomeric DNA. We consider this possibility unlikely as the Ku heterodimer binds
to telomeric DNA in vitro (Bianchi and de Lange, 1999)
and NHEJ-mediated joining of telomeric DNA ends has
326 Molecular Cell 26, 323–334, May 11, 2007 ª2007 Elsevier Inc.
Molecular Cell
Inhibition of NHEJ by Human RAP1/TRF2 Complex
been observed in cells expressing a dominant-negative
version of TRF2 (Smogorzewska et al., 2002). Second,
telomeric DNA may be specifically processed into a structure that is incompatible with ligation, such as long singlestranded overhangs. Instead or in addition, the G-rich
strand may adopt G-quadruplex structures that would
render the 30 end inaccessible to ligation. We can exclude
the possibility that exonucleolytic degradation of the
C-rich strand generated long 30 overhangs, as the radiolabeled terminal 50 phosphate was retained equally well on
telomeric and nontelomeric ends (e.g., Figure 1B). Furthermore, the ability of T4 DNA ligase to join telomeric
ends was only slightly impaired by prior or simultaneous
incubation with human cell-free extract (Figure 2D and
data not shown). It therefore appears unlikely that telomeric DNA end processing or structural modifications
account for the protection from NHEJ in this assay.
Further characterization of telomere-specific inhibition
of NHEJ revealed that end-joining was blocked even when
the terminal sequence was nontelomeric (Figure 2E). Inhibition was as efficient whether the DNA substrate terminated in telomeric repeats (lane 5) or whether 25 nontelomeric nucleotides were present between the telomeric
repeats and the 30 end (lane 7). Addition of 44 nucleotides
of nontelomeric sequence restored end-joining to the
same level as seen in the absence of telomeric sequence
(compare lanes 4 and 8). The protective effect exerted by
telomeric repeats therefore extends over some distance,
suggesting that sequence-specific end-binding is not
required for telomere capping.
In vivo, all telomeres have the same polarity with the
G-rich sequence representing the strand that contributes
the 30 terminus at each chromosome end. Placing the
TTAGGG sequence at the 30 end of the end-joining substrate therefore mimics a natural chromosome end,
whereas a 50 -TTAGGG end would only be found in cells
as the consequence of a DNA break proximal to a telomere
(Figure 2F). A comparison of the two possible orientations
revealed that only the TTAGGG-30 orientation provided
protection from NHEJ, whereas a 50 -TTAGGG terminus
was joint as efficiently as a nontelomeric substrate (Figure 2G). Protection of telomeric DNA ends from NHEJ
in vitro therefore has a defined polarity consistent with
the protection of natural chromosome termini.
TRF2, but Not TRF1 or Ku, Protects Telomere Ends
Several proteins have been implicated in chromosome
capping based on chromosomal aberrations observed
in mutant cells or as a result of RNAi-mediated knockdown. We were particularly interested in the role of Ku in
maintaining telomere integrity, as telomere fusions have
been reported for Ku70 and Ku80 mouse knockout cells
(d’Adda di Fagagna et al., 2001; Hsu et al., 2000; Samper
et al., 2000). Immunodepletion of Ku by 80% (Figure 3A,
lane 2) abolished NHEJ-mediated fusions of nontelomeric
ends (Figure 3B, lane 7). However, no fusions between
telomeric ends were detected either (lane 2). Instead, we observed reduced protection of all DNA ends as evidenced
by the loss of terminal radiolabel from the DNA substrates
(Figure 3B, compare signals for monomers). These observations are consistent with a model in which Ku protects
DNA ends at break sites as well as chromosome termini.
In the absence of Ku, another DNA repair pathway not reconstituted in this assay may be engaged and account for
telomere fusions observed in Ku/ cells in vivo.
Depletion of the telomere repeat binding factors TRF1
and TRF2 (Figure 3A, lanes 4 and 5) resulted in loss of telomere protection. Using fractions in which TRF1 and TRF2
were reduced to below the limit of detection (<2%), telomeric ends were joined as efficiently as nontelomeric ends
(Figure 3B, compare lanes 4 and 5 with lanes 9 and 10).
Depletion with an antibody raised against human RAP1 resulted in the concomitant loss of TRF1 and TRF2, consistent with the idea that these proteins interact with each
other as part of the shelterin complex (Figure 3A, lane 3).
Again, the depleted fraction failed to protect telomeric
ends and supported efficient ligation of all substrates (Figure 3B, lanes 3 and 8). Robust end-joining was also observed with DNA substrates containing telomeric repeats
at both ends (Figure S1A, lane 5). We surmised that our
in vitro system recapitulates the requirement for TRF2 in
chromosome end protection.
It is a matter of great interest whether end protection in
cells is mediated by TRF2 alone, the shelterin complex,
or TRF2 in association with other factors. To address this
issue biochemically we expressed and purified recombinant versions of TRF1 and TRF2 (Figure S2B). Telomeric
DNA-binding activity was verified for both proteins in electrophoretic mobility shift assays (data not shown). Despite
efficient binding to the substrate DNA, the addition of up to
150 fmol of TRF1 did not impair end-joining of telomeric
substrates (Figure 3C). In contrast, addition of only
15 fmol of TRF2 inhibited joining reactions involving telomeric ends (Figure 3D). Quantitative western analysis
showed that this amount of TRF2 corresponds to the
amount of endogenous TRF2 present in untreated cell extract (Figure S2C). Despite the ability of TRF2 to also interact with nontelomeric DNA (Amiard et al., 2007; our unpublished data), adding TRF2 had no effect on end-joining of
nontelomeric control substrates (Figure 3D, lanes 7–10).
TRF2 Is Required, but Not Sufficient,
for Telomere Protection
TRF2-depleted fractions also showed a reduced level of
RAP1 (Figure 3A, lanes 4 and 5). This is consistent with
the idea that RAP1 is recruited to telomeres via a direct
interaction with TRF2 (Li et al., 2000). Depletion of TRF2
reduced the level of RAP1 by 50%, suggesting that not
all RAP1 exists in a complex with TRF2 or that the RAP1TRF2 interaction is disrupted during immunoprecipitation.
In agreement with the former possibility and the proposed
1:1 stoichiometry (Zhu et al., 2000), a molar excess of
RAP1 over TRF2 was detected in cell extracts (Figures
S2C and S2D). When RAP1 was immunodepleted, RAP1
and TRF2 levels were reduced by more than 90% (Figure 3A, lane 3).
Molecular Cell 26, 323–334, May 11, 2007 ª2007 Elsevier Inc. 327
Molecular Cell
Inhibition of NHEJ by Human RAP1/TRF2 Complex
Figure 3. Telomeric Proteins Mediate Protection of TTAGGG Repeats from NHEJ
(A) Immunodepletion of Ku, RAP1, TRF1, and TRF2 verified by western analysis. Antibodies used for immunodepletion are listed above each lane. The
antibody against TRF1 (lane 5) was found to crossreact with TRF2 (see Figure S2A). Rabbit IgG was used as depletion control and RAD50 as loading
control.
(B) End-joining reactions using immunodepleted fractions (15 mg) shown in (A) and BbsI-digested pNB146.
(C) Indicated amounts of purified recombinant TRF1 were added to TRF1/2-depleted fraction (15 mg) prior to the addition of DNA substrate.
(D) As in (C) except that purified recombinant TRF2 was added.
Fractions depleted of RAP1 (and thereby TRF2) supported efficient end-joining of telomeric termini (Figure 4A,
lane 2). Interestingly though, the addition of recombinant
TRF2 to endogenous levels did not restore the protection
of telomeric DNA ends in RAP1-depleted fractions (lanes
3–6). Telomere protection was apparently restored when
recombinant TRF2 was added at a 200-fold higher concentration (Figure 4B, lane 4). However, under these conditions the specific polarity of protection was lost and
end-joining was also inhibited at 50 telomeric DNA termini,
suggesting a nonspecific sequestration of DNA substrates
(lane 8). In summary, these results suggest that specific
protection of 30 telomeric ends from NHEJ requires more
than just TRF2.
To identify factors that may cooperate with TRF2 in mediating the protection of 30 telomeric ends, we purified
TRF2 from HeLa cell nuclear extract (Figure 5A). The elution profile of the final gel filtration step revealed that
TRF2 was part of a complex with an apparent molecular
weight of 650 kDa (Figure 5B). In contrast, purified recombinant TRF2 was found to elute in a single peak at
an apparent molecular weight of 440 kDa. TRF2 has previously been reported to be a dimer in solution but elute at
a much higher apparent molecular weight on gel filtration
due to the presence of a long unstructured linker in each
subunit (Court et al., 2005). A nonglobular shape of TRF2
was also supported by sucrose gradient centrifugation
analysis (our unpublished data).
Unlike recombinant TRF2, the endogenous TRF2 complex efficiently restored protection of 30 telomeric ends
from NHEJ (Figure 5C). Analysis by western blotting and
mass spectrometry revealed that the 650 kDa TRF2
complex contained RAP1, but not TRF1, TIN2, or POT1
(Figure 5B and data not shown). When recombinant RAP1
was purified (Figure S2B) and incubated with TRF2 a stable
complex was formed that was easily separated from excess TRF2 by anion exchange chromatography (data not
shown). This reconstituted RAP1/TRF2 complex eluted in
two overlapping peaks from a Superdex 200 gel filtration
column, with the major peak coinciding with the elution
profile of the endogenous TRF2 complex (Figure 5B). Although we cannot exclude the possibility that elution of the
recombinant and endogenous TRF2 complexes in equivalent fractions is coincidental, the results suggest that the
endogenous TRF2 complex may be solely comprised of
RAP1 and TRF2.
Addition of recombinant RAP1 to RAP1-depleted endjoining reactions did not restore the protection of telomeric
ends (Figure 6A). However, when purified RAP1 and TRF2
were added together, the telomeric ends were again protected from NHEJ (Figure 6B, lanes 7–9). Like the effect
observed with endogenous factors, protection by the reconstituted RAP1/TRF2 complex was highly directional,
affecting only DNA ends where the G-rich strand contributes the 30 terminus (Figure 6C). Taken together, our data
support a critical role for a RAP1/TRF2 complex in
328 Molecular Cell 26, 323–334, May 11, 2007 ª2007 Elsevier Inc.
Molecular Cell
Inhibition of NHEJ by Human RAP1/TRF2 Complex
Figure 4. TRF2 Is Not Sufficient to Protect Telomeric Ends
(A) Recombinant TRF2 was added to extract depleted with anti-human
RAP1 as shown in Figure 3A, lane 3, followed by the addition of BbsIcut pNB146.
(B) End-joining assays were carried out with IgG (M)- or hRAP1 (R)-depleted extract and DNA substrates containing 24 telomeric repeats at
the 30 or 50 end, respectively.
capping short telomeres and protecting them from endto-end fusions.
DISCUSSION
The finding that a biochemical assay for DNA repair by
NHEJ also recapitulates telomere capping provides an
opportunity to gain mechanistic insights into chromosome
end protection. We have shown here that the efficient protection of telomeric DNA ends from NHEJ requires a complex comprised of RAP1 and TRF2. While TRF2 has previously been implicated in telomere capping, this work
provides evidence for a direct role of human RAP1 in the
protection of telomeres from NHEJ. It is presently unknown whether other factors from the end-joining fraction
cooperate with the RAP1/TRF2 complex in the protection
of telomeric ends.
Perhaps our most unexpected finding is that a relatively
small number of telomeric repeats are sufficient to prevent
NHEJ from acting on a nearby DNA end. Human telomeres
average over 1000 TTAGGG repeats, but NHEJ-mediated
fusions were undetectable in the presence of 12 repeats
(Figure 2C). This observation is at odds with a model of
telomere protection in which t loop structures are required
to protect chromosome ends from NHEJ, as the telomeres
used here are far too short for loop formation. The finding
that such short telomeres are nevertheless protected from
NHEJ indicates that cells may use more than one pathway
to safeguard chromosome ends. Analyses of mutant and
knockout cell lines as well as RNAi experiments have implicated numerous factors in the protection of chromosome ends including TRF1, TRF2, POT1, Ku, DNA-PKcs,
Artemis, Apollo, PARP, and RAD50 (Ferreira et al., 2004;
Lenain et al., 2006; van Overbeek and de Lange, 2006).
The involvement of so many proteins suggests that multiple pathways cooperate to ensure the integrity of all chromosome ends despite dynamic differences in telomere
length and structure. Notably, STELA revealed that a subset of human telomeres is indeed as short as eight repeats
(Baird et al., 2003). We propose that these extremely short
telomeres may be protected from undergoing NHEJ-mediated fusion by the RAP1/TRF2-dependent mechanism
described here. In addition, the same mechanism may
also be responsible for the protection of longer telomeres
when they are in an open conformation, such as immediately after replication.
TRF2 seems to play a central role in protecting the majority of telomeres in mouse and human cells. This has in
part been attributed to its t loop promoting activity (Stansel et al., 2001). In addition, TRF2 has been shown to bind
the ATM kinase and block its activation, a mechanism by
which a DNA damage response may be inhibited locally at
telomeres where TRF2 is abundant (Karlseder et al., 2004).
This raises the interesting question whether TRF2 partially
compromises the repair of breaks in subtelomeric DNA by
locally weakening the DNA damage response. Consistent
with such an effect, studies in budding yeast showed that
a checkpoint response triggered by a DSB is inhibited in
the vicinity of telomeric repeats (Michelson et al., 2005).
Collectively, these findings predict that subtelomeric DNA
may be particularly vulnerable to DNA damage. It is important to note in this context that the mechanism of telomere protection described here shows a strong polarity,
whereby end-joining is only inhibited at DNA ends 30 to
the TTAGGG repeats. As the G-rich strand is oriented in
the 50 to 30 direction on all naturally occurring telomeres,
this mechanism of telomere capping would not interfere
with the repair of breaks upstream of telomeric repeats.
The strict polarity in end protection is surprising in light
of what is known about the DNA-binding mode of TRF2. A
binding site for a TRF2 homodimer contains two YTAGGG
TTR half-sites, each bound by one homeodomain (Broccoli et al., 1997; Court et al., 2005). As each DNA-binding
domain is connected to the dimerization domain by a long,
highly flexible linker, DNA-binding affinity is thought to be
unaffected by the relative orientation and, to a certain extent, spacing of the two half-sites (Fairall et al., 2001). It
should therefore be expected that a tandem array of 12
telomeric repeats is bound equally well near the 50 or 30
end of a DNA substrate. Indeed, gel retardation assays
confirmed that TRF2 and the RAP1/TRF2 complex bind
both substrates (our unpublished data), yet inhibition of
Molecular Cell 26, 323–334, May 11, 2007 ª2007 Elsevier Inc. 329
Molecular Cell
Inhibition of NHEJ by Human RAP1/TRF2 Complex
Figure 5. A TRF2-Containing Complex Reconstitutes Protection from NHEJ
(A) Purification scheme for TRF2 from HeLa cells.
(B) Gel filtration analysis of recombinant TRF2 (top), TRF2 complex from HeLa cells (middle), and reconstituted RAP1/TRF2 complex (bottom) on
Superdex 200 PC3.2/30. The absorbance at 280 nm is plotted over the elution volume. Fractions containing TRF2 and RAP1 were identified by western blotting (middle) or Coomassie staining (bottom). Molecular weight markers are albumin (68 kDa), catalase (232 kDa), ferritin (440 kDa), and
thyroglobulin (669 kDa).
(C) Endogenous TRF2 complex purified from HeLa cells restores telomere protection. Approximately 0.25 ng (lane 3), 0.5 ng (lane 4), and 1 ng (lane 5)
of RAP1/TRF2 complex were used in the end-joining assay with RAP1-depleted fraction and EcoRI-cut pNB146.
NHEJ was only observed when a DNA end was present on
the 30 site of the TTAGGG repeats (Figure 2G).
A preference for TRF2 to bind DNA ends has previously
been reported (Stansel et al., 2001), but this study did not
address whether TRF2 preferentially bound a 30 over a 50
telomeric end. Rather, terminal binding was found to be
dependent on the presence of a telomeric double-strand
to single-strand transition, a property that is at least in
part mediated by the N-terminal basic domain of TRF2
(Fouche et al., 2006). We found that inhibition of NHEJ
did not depend on the presence of a 30 single-stranded
overhang, nor did it require telomeric repeats directly adjacent to the DNA end. When 25 bp of nontelomeric DNA
were placed 30 to 12 telomeric repeats the inhibitory effect
on NHEJ was not diminished. Collectively, these results
suggest that telomere protection by the RAP1/TRF2 complex must involve a directional component distinct from 30
end recognition of the G strand.
330 Molecular Cell 26, 323–334, May 11, 2007 ª2007 Elsevier Inc.
Molecular Cell
Inhibition of NHEJ by Human RAP1/TRF2 Complex
Figure 6. Protection from NHEJ by Recombinant RAP1/TRF2
(A) Indicated amounts of purified recombinant RAP1 were added to RAP1-depleted fraction followed by the addition of BamHI-digested pNB146
(24 telomeric repeats near the 30 end).
(B) Recombinant RAP1 and TRF2 were added to TRF2- or RAP1-depleted fractions.
(C) Reconstituted and purified RAP1/TRF2 complex was added to RAP1-depleted fractions prior to the addition of DNA substrates with 24 telomeric
repeats at the 50 or 30 end, respectively. The amount of RAP1/TRF2 complex added is given in nanograms, as the exact stoichiometry of RAP1 and
TRF2 in the complex is presently unknown.
The removal of TRF2 from telomeres results in a dramatic
increase in chromosome fusions (van Steensel et al.,
1998). As TRF2 recruits RAP1 to telomeres, RAP1 would
have been lost alongside TRF2 in these experiments. A
general role for RAP1 in protecting telomeres from NHEJ
is supported by studies in yeast. In Schizosaccharomyces
pombe, ligase IV-dependent chromosome fusions accumulate in cells lacking rap1+ or taz1+, a TRF1/2 ortholog
(Miller et al., 2005). These fusions occur specifically in G1
when NHEJ is the dominant repair pathway. In Saccharomyces cerevisiae, a Taz1/TRF2-related protein is not involved in telomere protection. Instead, RAP1 binds directly
to telomeric sequences (Conrad et al., 1990; Konig et al.,
1996) where it functions in telomere length regulation
(Marcand et al., 1997) and prevents telomere fusions by
NHEJ (Pardo and Marcand, 2005). The conservation of
RAP1 as a telomeric protein and as a negative regulator
of telomere length across species suggests that a general
role in NHEJ suppression may also be universal. Future
experiments will need to determine whether the human
RAP1/TRF2 complex specifically protects very short telomeres or is important for the protection of all telomeres
from NHEJ.
EXPERIMENTAL PROCEDURES
Cell Lines and Extract Preparation
GM00558 cells were obtained from the NIGMS Human Genetic Mutant
Cell Repository (Camden, NJ), and HeLa S3 cells were from ATCC.
GM00558 cells were grown in RPMI1640 supplemented with 15%
heat-inactivated fetal bovine serum and 2 mM L-glutamine. A detailed
protocol for the preparation of end-joining-competent extract can be
found as Supplemental Data online. HeLa cells were grown in suspension to 9 3 105 cells/ml in Joklik minimal essential medium supplemented with 10% fetal calf serum, 2 mM L-glutamine, and nonessential amino acids.
DNA and Proteins
Between 2 and 24 telomeric and scrambled repeats were introduced
into the multicloning sites of pBluescript and pDEA-7Z, and constructs
were sequence verified. The pDEA-7Z-derived plasmid containing
(TTAGGG)24 is referred to as pNB146 in the figure legends. Plasmids
were linearized with BbsI, EcoRI, BamHI, or NotI to generate substrates with telomeric repeats at or near the 30 end of the G-rich strand.
Digestion with BsmBI, HindIII, and XhoI gave rise to DNA substrates
with the G-rich strand at or near the 50 end. Linearized plasmids
were dephosphorylated using calf intestine phosphatase and 32P-labeled with polynucleotide kinase in the presence of [g-32P]ATP.
TRF2 was purified from 4.4 3 1010 HeLa cells at 4 C as follows. For
each of four batches 1.1 3 1010 cells were harvested by centrifugation,
Molecular Cell 26, 323–334, May 11, 2007 ª2007 Elsevier Inc. 331
Molecular Cell
Inhibition of NHEJ by Human RAP1/TRF2 Complex
washed in phosphate-buffered saline, and resuspended in five packed
cell volumes (PCV) of buffer A (10 mM HEPES [pH 7.9]; 1.5 mM MgCl2;
10 mM KCl; 0.5 mM dithiothreitol; 1 mM phenylmethylsulfonyl fluoride;
1 mg/ml of pepstatin, chymostatin, and leupeptin; and 7.5 units/l aprotinin), then incubated for 15 min on ice and collected by centrifugation.
Cells were resuspended in 2 PCV of buffer A and lysed by Dounce homogenization using a loose-fitting pestle. The nuclei were collected by
centrifugation, resuspended in 27.5 ml of buffer C (20 mM HEPES
[pH 7.9], 25% glycerol, 1.5 mM MgCl2, 0.2 mM EDTA, 0.5 mM dithiothreitol, and protease inhibitors as for buffer A) and lysed by Dounce
homogenization using a tight-fitting pestle. Sodium chloride was added
to 0.42 M, followed by incubation on ice for 30 min. The soluble fraction
was recovered by centrifugation, and ammonium sulfate was added
slowly to 20% (w/v), followed by stirring on ice for 60 min. The soluble
fraction was recovered by centrifugation and subjected to a second
round of ammonium sulfate precipitation at 60% (NH4)2SO4. Precipitated proteins were recovered by centrifugation and frozen at 80 C
until use. Four pellets were combined for subsequent steps, solubilized in PK1 buffer (50 mM Tris-HCl [pH 8.0], 0.1 M KCl, 5% glycerol,
1 mM EDTA, 0.02% Tween 20, and 1 mM dithiothreitol), and dialyzed
for 2.5 hr against the same buffer. The sample was cleared by centrifugation and applied onto a phosphocellulose column (Whatman
P11) equilibrated with PK1. Proteins were eluted with a KCl step gradient in 0.1 M increments in the same buffer, and TRF2-containing fractions were identified by western blotting and pooled. This fraction was
dialyzed against buffer T1 (25 mM HEPES [pH 7.5], 50 mM KCl, 5%
glycerol, 0.5 mM EDTA, and 0.5 mM dithiothreitol) and applied to a
DEAE column (Bio-Rad) equilibrated in buffer T1, and proteins were
eluted with a step gradient as above. TRF2-containing fractions were
diluted with 2 vol buffer T2 (20 mM potassium phosphate [pH 6.8],
5% glycerol, 0.1 mM EDTA, and 0.5 mM dithiothreitol) and applied to
a monoS HR5/5 column (GE Healthcare). A linear gradient from 0 to
0.5 M KCl was applied, and TRF2-containing fractions (0.35 M
KCl) were dialyzed against T3 buffer (250 mM Tris-HCl [pH 8.0], 10%
glycerol, 0.1 M KCl, 0.5 mM EDTA, and 0.5 mM dithiothreitol) and
applied to a miniQ PC3.2/3 column on a SMART system (Pharmacia).
A linear gradient from 0 to 0.7 M KCl was applied and TRF2 eluted
around 0.35 M KCl. This fraction was either used directly in endjoining assays or subjected to size fractionation on a Superdex 200
PC3.2/30 column.
His6-tagged TRF1 and TRF2 were expressed in baculovirus infected
Sf21 cells and were purified on Ni-NTA resin (QIAGEN) and monoQ (GE
Healthcare). GST-tagged human RAP1 was expressed in E. coli and
purified on glutathione beads, followed by cleavage with thrombin
and further purification on monoQ. For the purification of the recombinant RAP1/TRF2 complex, cleaved GST-RAP1 (7.2 mg) was incubated
with TRF2 (7.5 mg) in 20 mM Tris (pH 8.0), 10% glycerol, 0.1 M KCl, and
2.5 mM b-mercaptoethanol for 3 hr at 4 C. The RAP1/TRF2 complex
was then separated from uncomplexed TRF2, GST, and RAP1 degradation products on a MonoQ HR 5/5 column. Gel filtration analysis was
carried out on a SMART system using a Superdex 200 PC3.2/30
column.
IgG control. After 4 hr at 4 C supernatants were recovered, aliquoted,
and frozen at 80 C until use. The following antibodies were used:
XRCC4 (ab145; Abcam) and Ku70/80 (ab3108; Abcam), Ku86 (Santa
Cruz Biotechnology), hRAP1 (BL735; Bethyl Laboratories), TRF1
(TRF-78; Sigma), TRF2 (4A794; Upstate), and Rad50 (13B3; GeneTex).
For western blotting all antibodies were used at a dilution of 1:1000.
End-Joining Assay
Standard reactions (10 ml) contained 5 or 10 ng of EcoRI-cut DNA substrate and 15 mg of GM00558 extract. A detailed protocol is available
online as Supplemental Data. For experiments with inhibitors or blocking antibodies, the extract was preincubated with the blocking agent
on ice for 60 min prior to a shift to 37 C and the addition of DNA
substrate.
Bianchi, A., and de Lange, T. (1999). Ku binds telomeric DNA in vitro.
J. Biol. Chem. 274, 21223–21227.
Immunodepletion and Western Analysis
Immunodepletions were carried out in 130 ml reactions containing protein A/G beads (Calbiochem; 30 ml), 150 mg GM00558 extract in EJ
buffer (50 mM tri-ethanol amine/acetic acid [pH 7.5], 60 mM potassium
acetate, 0.5 mM Mg[OAc]2, 1 mM ATP, 1 mM dithiothreitol, and
0.1 mg/ml bovine serum albumin), and 2 mg of purified antibody or
Supplemental Data
Supplemental Data include two figures, Supplemental Experimental
Procedures, and Supplemental References and can be found with
this article online at http://www.molecule.org/cgi/content/full/26/3/
323/DC1/.
ACKNOWLEDGMENTS
We thank Jay Sarthy for critical reading of the manuscript, Titia de
Lange for TRF1 and TRF2 containing baculovirus, Songyang Zhou
for the human RAP1 expression construct, and Jon Woods and
Howard Cook for telomeric DNA substrates. We are indebted to the
Stowers Institute Molecular Biology Facility and Proteomics Center
for excellent assistance. This work was supported by a Basil O’Connor
Award to P.B. and by the Pew Scholars Program in Biomedical
Sciences and the Stowers Institute for Medical Research.
Received: February 26, 2007
Revised: March 22, 2007
Accepted: March 30, 2007
Published: May 10, 2007
REFERENCES
Amiard, S., Doudeau, M., Pinte, S., Poulet, A., Lenain, C., FaivreMoskalenko, C., Angelov, D., Hug, N., Vindigni, A., Bouvet, P., et al.
(2007). A topological mechanism for TRF2-enhanced strand invasion.
Nat. Struct. Mol. Biol. 14, 147–154.
Ancelin, K., Brunori, M., Bauwens, S., Koering, C.E., Brun, C., Ricoul,
M., Pommier, J.P., Sabatier, L., and Gilson, E. (2002). Targeting assay
to study the cis functions of human telomeric proteins: evidence for
inhibition of telomerase by TRF1 and for activation of telomere degradation by TRF2. Mol. Cell. Biol. 22, 3474–3487.
Baird, D.M., Rowson, J., Wynford-Thomas, D., and Kipling, D. (2003).
Extensive allelic variation and ultrashort telomeres in senescent human
cells. Nat. Genet. 33, 203–207.
Baumann, P., and West, S.C. (1998). DNA end-joining catalyzed by
human cell-free extracts. Proc. Natl. Acad. Sci. USA 95, 14066–14070.
Baumann, P., and Cech, T.R. (2001). Pot1, the putative telomere endbinding protein in fission yeast and humans. Science 292, 1171–1175.
Baumann, P., Podell, E.R., and Cech, T.R. (2002). Human Pot1 (protection of telomeres) protein: cytolocalization, gene structure, and alternative splicing. Mol. Cell. Biol. 22, 8079–8087.
Bilaud, T., Brun, C., Ancelin, K., Koering, C.E., Laroche, T., and Gilson,
E. (1997). Telomeric localization of TRF2, a novel human telobox
protein. Nat. Genet. 17, 236–239.
Bradshaw, P.S., Stavropoulos, D.J., and Meyn, M.S. (2005). Human telomeric protein TRF2 associates with genomic double-strand breaks
as an early response to DNA damage. Nat. Genet. 37, 193–197.
Broccoli, D., Smogorzewska, A., Chong, L., and de Lange, T. (1997).
Human telomeres contain two distinct Myb-related proteins, TRF1
and TRF2. Nat. Genet. 17, 231–235.
Budman, J., and Chu, G. (2005). Processing of DNA for nonhomologous end-joining by cell-free extract. EMBO J. 24, 849–860.
332 Molecular Cell 26, 323–334, May 11, 2007 ª2007 Elsevier Inc.
Molecular Cell
Inhibition of NHEJ by Human RAP1/TRF2 Complex
Burma, S., Chen, B.P., and Chen, D.J. (2006). Role of non-homologous
end joining (NHEJ) in maintaining genomic integrity. DNA Repair
(Amst.) 5, 1042–1048.
Houghtaling, B.R., Cuttonaro, L., Chang, W., and Smith, S. (2004). A
dynamic molecular link between the telomere length regulator TRF1
and the chromosome end protector TRF2. Curr. Biol. 14, 1621–1631.
Cech, T.R. (2004). Beginning to understand the end of the chromosome. Cell 116, 273–279.
Hsu, H.L., Gilley, D., Galande, S.A., Hande, M.P., Allen, B., Kim, S.H.,
Li, G.C., Campisi, J., Kohwi-Shigematsu, T., and Chen, D.J. (2000). Ku
acts in a unique way at the mammalian telomere to prevent end joining.
Genes Dev. 14, 2807–2812.
Celli, G.B., and de Lange, T. (2005). DNA processing is not required
for ATM-mediated telomere damage response after TRF2 deletion.
Nat. Cell Biol. 7, 712–718.
Chappell, C., Hanakahi, L.A., Karimi-Busheri, F., Weinfeld, M., and
West, S.C. (2002). Involvement of human polynucleotide kinase in double-strand break repair by non-homologous end joining. EMBO J. 21,
2827–2832.
Chong, L., van Steensel, B., Broccoli, D., Erdjument-Bromage, H.,
Hanish, J., Tempst, P., and de Lange, T. (1995). A human telomeric
protein. Science 270, 1663–1667.
Conrad, M.N., Wright, J.H., Wolf, A.J., and Zakian, V.A. (1990). RAP1
protein interacts with yeast telomeres in vivo: overproduction alters
telomere structure and decreases chromosome stability. Cell 63,
739–750.
Court, R., Chapman, L., Fairall, L., and Rhodes, D. (2005). How the
human telomeric proteins TRF1 and TRF2 recognize telomeric DNA:
a view from high-resolution crystal structures. EMBO Rep. 6, 39–45.
d’Adda di Fagagna, F., Hande, M.P., Tong, W.M., Roth, D., Lansdorp,
P.M., Wang, Z.Q., and Jackson, S.P. (2001). Effects of DNA nonhomologous end-joining factors on telomere length and chromosomal stability in mammalian cells. Curr. Biol. 11, 1192–1196.
d’Adda di Fagagna, F., Teo, S.H., and Jackson, S.P. (2004). Functional
links between telomeres and proteins of the DNA-damage response.
Genes Dev. 18, 1781–1799.
de Lange, T. (2004). T-loops and the origin of telomeres. Nat. Rev. Mol.
Cell Biol. 5, 323–329.
de Lange, T. (2005). Shelterin: the protein complex that shapes and
safeguards human telomeres. Genes Dev. 19, 2100–2110.
Fairall, L., Chapman, L., Moss, H., de Lange, T., and Rhodes, D. (2001).
Structure of the TRFH dimerization domain of the human telomeric
proteins TRF1 and TRF2. Mol. Cell 8, 351–361.
Fouche, N., Cesare, A.J., Willcox, S., Ozgur, S., Compton, S.A., and
Griffith, J.D. (2006). The basic domain of TRF2 directs binding to
DNA junctions irrespective of the presence of TTAGGG repeats. J.
Biol. Chem. 281, 37486–37495.
Ferreira, M.G., and Cooper, J.P. (2004). Two modes of DNA doublestrand break repair are reciprocally regulated through the fission yeast
cell cycle. Genes Dev. 18, 2249–2254.
Ferreira, M.G., Miller, K.M., and Cooper, J.P. (2004). Indecent exposure: when telomeres become uncapped. Mol. Cell 13, 7–18.
Griffith, J.D., Comeau, L., Rosenfield, S., Stansel, R.M., Bianchi, A.,
Moss, H., and de Lange, T. (1999). Mammalian telomeres end in a large
duplex loop. Cell 97, 503–514.
Hanakahi, L.A., Bartlet-Jones, M., Chappell, C., Pappin, D., and West,
S.C. (2000). Binding of inositol phosphate to DNA-PK and stimulation
of double-strand break repair. Cell 102, 721–729.
He, H., Multani, A.S., Cosme-Blanco, W., Tahara, H., Ma, J., Pathak,
S., Deng, Y., and Chang, S. (2006). POT1b protects telomeres from
end-to-end chromosomal fusions and aberrant homologous recombination. EMBO J. 25, 5180–5190.
Hockemeyer, D., Sfeir, A.J., Shay, J.W., Wright, W.E., and de Lange, T.
(2005). POT1 protects telomeres from a transient DNA damage
response and determines how human chromosomes end. EMBO J.
24, 2667–2678.
Hockemeyer, D., Daniels, J.P., Takai, H., and de Lange, T. (2006). Recent expansion of the telomeric complex in rodents: two distinct POT1
proteins protect mouse telomeres. Cell 126, 63–77.
Huang, J., and Dynan, W.S. (2002). Reconstitution of the mammalian
DNA double-strand break end-joining reaction reveals a requirement
for an Mre11/Rad50/NBS1-containing fraction. Nucleic Acids Res.
30, 667–674.
Hug, N., and Lingner, J. (2006). Telomere length homeostasis. Chromosoma 115, 413–425.
Iwano, T., Tachibana, M., Reth, M., and Shinkai, Y. (2003). Importance
of TRF1 for functional telomere structure. J. Biol. Chem. 279, 1142–
1148.
Karlseder, J., Broccoli, D., Dai, Y., Hardy, S., and de Lange, T. (1999).
p53- and ATM-dependent apoptosis induced by telomeres lacking
TRF2. Science 283, 1321–1325.
Karlseder, J., Kachatrian, L., Takai, H., Mercer, K., Hingorani, S.,
Jacks, T., and de Lange, T. (2003). Targeted deletion reveals an essential function for the telomere length regulator Trf1. Mol. Cell. Biol. 23,
6533–6541.
Karlseder, J., Hoke, K., Mirzoeva, O.K., Bakkenist, C., Kastan, M.B.,
Petrini, J.H., and de Lange, T. (2004). The telomeric protein TRF2 binds
the ATM kinase and can inhibit the ATM-dependent DNA damage
response. PLoS Biol. 2, E240.
Kim, S.H., Kaminker, P., and Campisi, J. (1999). TIN2, a new regulator
of telomere length in human cells. Nat. Genet. 23, 405–412.
Konig, P., Giraldo, R., Chapman, L., and Rhodes, D. (1996). The crystal
structure of the DNA-binding domain of yeast RAP1 in complex with
telomeric DNA. Cell 85, 125–136.
Lenain, C., Bauwens, S., Amiard, S., Brunori, M., Giraud-Panis, M.J.,
and Gilson, E. (2006). The Apollo 50 exonuclease functions together
with TRF2 to protect telomeres from DNA repair. Curr. Biol. 16, 1303–
1310.
Li, B., Oestreich, S., and de Lange, T. (2000). Identification of human
Rap1: implications for telomere evolution. Cell 101, 471–483.
Liu, D., Safari, A., O’Connor, M.S., Chan, D.W., Laegeler, A., Qin, J.,
and Songyang, Z. (2004). PTOP interacts with POT1 and regulates
its localization to telomeres. Nat. Cell Biol. 6, 673–680.
Loayza, D., and de Lange, T. (2003). POT1 as a terminal transducer of
TRF1 telomere length control. Nature 423, 1013–1018.
Lundblad, V. (2002). Telomere maintenance without telomerase.
Oncogene 21, 522–531.
Lustig, A.J. (2003). Clues to catastrophic telomere loss in mammals
from yeast telomere rapid deletion. Nat. Rev. Genet. 4, 916–923.
Marcand, S., Gilson, E., and Shore, D. (1997). A protein-counting
mechanism for telomere length regulation in yeast. Science 275,
986–990.
Michelson, R.J., Rosenstein, S., and Weinert, T. (2005). A telomeric
repeat sequence adjacent to a DNA double-stranded break produces
an anticheckpoint. Genes Dev. 19, 2546–2559.
Miller, K.M., and Cooper, J.P. (2003). The telomere protein taz1 is
required to prevent and repair genomic DNA breaks. Mol. Cell 11,
303–313.
Miller, K.M., Ferreira, M.G., and Cooper, J.P. (2005). Taz1, Rap1 and
Rif1 act both interdependently and independently to maintain telomeres. EMBO J. 24, 3128–3135.
Pardo, B., and Marcand, S. (2005). Rap1 prevents telomere fusions by
nonhomologous end joining. EMBO J. 24, 3117–3127.
Molecular Cell 26, 323–334, May 11, 2007 ª2007 Elsevier Inc. 333
Molecular Cell
Inhibition of NHEJ by Human RAP1/TRF2 Complex
Samper, E., Goytisolo, F.A., Slijepcevic, P., van Buul, P.P., and Blasco,
M.A. (2000). Mammalian Ku86 protein prevents telomeric fusions independently of the length of TTAGGG repeats and the G-strand overhang. EMBO Rep. 1, 244–252.
Smogorzewska, A., Karlseder, J., Holtgreve-Grez, H., Jauch, A., and
de Lange, T. (2002). DNA ligase IV-dependent NHEJ of deprotected
mammalian telomeres in G1 and G2. Curr. Biol. 12, 1635–1644.
Stansel, R.M., de Lange, T., and Griffith, J.D. (2001). T-loop assembly
in vitro involves binding of TRF2 near the 30 telomeric overhang. EMBO
J. 20, 5532–5540.
van Overbeek, M., and de Lange, T. (2006). Apollo, an Artemis-related
nuclease, interacts with TRF2 and protects human telomeres in
S phase. Curr. Biol. 16, 1295–1302.
Verdun, R.E., Crabbe, L., Haggblom, C., and Karlseder, J. (2005).
Functional human telomeres are recognized as DNA damage in G2
of the cell cycle. Mol. Cell 20, 551–561.
Wang, R.C., Smogorzewska, A., and de Lange, T. (2004). Homologous
recombination generates T-loop-sized deletions at human telomeres.
Cell 119, 355–368.
West, S.C. (2003). Molecular views of recombination proteins and their
control. Nat. Rev. Mol. Cell Biol. 4, 435–445.
Wu, L., Multani, A.S., He, H., Cosme-Blanco, W., Deng, Y., Deng, J.M.,
Bachilo, O., Pathak, S., Tahara, H., Bailey, S.M., et al. (2006). Pot1 deficiency initiates DNA damage checkpoint activation and aberrant
homologous recombination at telomeres. Cell 126, 49–62.
Wyman, C., and Kanaar, R. (2006). DNA double-strand break repair:
all’s well that ends well. Annu. Rev. Genet. 40, 363–383.
van Steensel, B., Smogorzewska, A., and de Lange, T. (1998). TRF2
protects human telomeres from end-to-end fusions. Cell 92, 401–413.
Yang, Q., Zheng, Y.L., and Harris, C.C. (2005). POT1 and TRF2 cooperate to maintain telomeric integrity. Mol. Cell. Biol. 25, 1070–1080.
Veldman, T., Etheridge, K.T., and Counter, C.M. (2004). Loss of hPot1
function leads to telomere instability and a cut-like phenotype. Curr.
Biol. 14, 2264–2270.
Ye, J.Z., Hockemeyer, D., Krutchinsky, A.N., Loayza, D., Hooper, S.M.,
Chait, B.T., and de Lange, T. (2004). POT1-interacting protein PIP1:
a telomere length regulator that recruits POT1 to the TIN2/TRF1 complex. Genes Dev. 18, 1649–1654.
Verdun, R.E., and Karlseder, J. (2006). The DNA damage machinery
and homologous recombination pathway act consecutively to protect
human telomeres. Cell 127, 709–720.
Zhu, X.-D., Kuster, B., Mann, M., Petrini, J.H.J., and de Lange, T.
(2000). Cell-cycle-regulated assocation of RAD50/MRE11/NBS1 with
TRF2 and human telomeres. Nat. Genet. 25, 347–352.
334 Molecular Cell 26, 323–334, May 11, 2007 ª2007 Elsevier Inc.