Laboratory for Cell Biology and Genetics


Telomeres, the nucleoprotein complexes at the natural ends of chromosomes, protect chromosome ends and ensure their complete replication. The loss of telomere protection is the cause of the premature aging symptoms associated with Dyskeratosis congenita and other telomeropathies. Telomere dysfunction also plays an important role in the early stages of cancer and the activation of a telomere maintenance system (telomerase or ALT) is a hallmark of human cancer. We study how the telomeric shelterin complex prevents the activation of DNA damage signaling pathways and blocks various forms of double-strand break (DSB) repair at chromosome ends. We also work on the role of telomeres in genome instability in cancer.



Telomeres and cancer: tumor suppression and genome instability
Relevant Publications

Although human telomeres can be maintained by telomerase, most normal human cells lack (sufficient) telomerase and experience gradual telomere shortening. Eventually, the shortened telomeres become too short to fulfill their protective function, resulting in a block to further proliferation and a finite replicative lifespan. This programmed proliferation barrier is thought to function as a tumor suppressor pathway. The critically short telomeres activate the DNA damage response (DDR), which induces cell cycle arrest and senescence or apoptosis, depending on the cellular context. However, incipient cancer cells that have lost the p53/Rb pathways, ignore the cell cycle arrest signals and their telomeres shorten further during additional cell divisions. Eventually, the burden of dysfunctional telomeres in such cells leads to telomere-telomere fusions and formation of dicentric chromosomes. This stage of tumor development is referred to as telomere crisis and is thought to be an important source of genome instability in cancer.  Eventually, patent cancer will only arise from cells that escape telomere crisis by upregulation of telomerase or activation of the ALT pathway. We are studying the genomic consequences of telomere crisis to understand how telomere shortening contributes to genome instability.

Our most recent work has illuminated that the persistent DNA damage signal associated with telomere dysfunction can drive endoreduplication and formation of tetraploid cells. Tetraploidization is a hallmark of a large fraction of human cancers and our work indicates that some of these cancer may have become tetraploid as a consequence of their past telomere crisis. We have also found that the dicentric chromosomes formed in telomere crisis can fuel chromothripsis (chromosome shattering) and kataegis (hyper mutation clusters). Unlike what was previously believed, we showed that dicentric chromosomes can persist through mitosis and form long chromatin bridges between daughter cells. These bridges are attacked by a cytoplasmic nuclease, TREX1, which accesses the bridge DNA after transient rupture of the nuclear envelope. TREX1 digestion leads to extensively fragmented ssDNA that joins the daughter nuclei after bridge resolution. The fragmented DNA is joined haphazardly leading to the chromothriptic patterns that are accompanied by kataegis due to APOBEC editing of the single stranded DNA.



How shelterin solves the end-protection problem
Relevant Publications

The multi-subunit shelterin complex is crucial for the protection of telomeres from the DNA damage response and regulates telomere maintenance by telomerase. Shelterin is composed of six proteins: TRF1, TRF2, Rap1, TIN2, TPP1, and POT1. TRF1 and TRF2 bind to the duplex telomeric repeat array and anchor shelterin on telomeres. POT1 binds to single-stranded TTAGGG repeats and is recruited to telomeres through its interaction with TPP1. TPP1 in turns binds to TIN2, which interacts with both TRF1 and TRF2. Due to its specificity for the sequence and structure of the telomeric DNA, shelterin accumulates at the ends of human and mouse chromosomes but does not bind to DNA ends elsewhere in the genome. Thus, shelterin constitutes a unique marker of telomeres that allows cells to distinguish natural chromosome ends from sites of DNA damage. Our main approach to understanding how shelterin solves the end-protection problem is to generate mouse cells from which individual shelterin proteins can be removed using inducible systems, e.g. Cre-mediated deletion. Mouse and human shelterin are nearly identical except for the presence of two POT1 proteins (POT1a and POT1b) in mouse shelterin.

Cre-mediated deletion of individual shelterin subunits showed that shelterin is highly compartmentalized such that distinct subunits are dedicated to different DDR pathways. TRF2 is critical for the repression of ATM signaling and prevents fusion of telomeres by the c-NHEJ and alt-NHEJ pathways. POT1 is needed to prevent the activation of the ATR kinase at telomeres. POT1a is primarily responsible for this function, whereas POT1b controls the formation of the 3’ overhang. TPP1 mediates the functions of POT1 by tethering POT1 via TIN2 to the rest of shelterin. The function of Rap1 is to repress Homology-Directed Repair (HDR) together with one of the two POT1 proteins. TRF1 has no direct role in protecting the chromosome end but is dedicated to promoting the replication of the telomeric DNA.



Repression of ATM signaling and NHEJ by TRF2: t-loops
Relevant Publications

Inappropriate NHEJ at telomeres can lead to unstable dicentric chromosomes and needs to be stringently repressed. NHEJ-mediated fusion of telomeres is rampant when TRF2 is deleted from mouse cells, resulting in long trains of joined chromosomes. The telomere fusions that occur in the absence of TRF2 are formed through the loading of the Ku70/80 heterodimer onto telomere ends and involve ligation by DNA ligase IV, indicating that they are due to c-NHEJ. Alternative-NHEJ, mediated by PARP1 and DNA ligase III can also take place at telomeres but only when shelterin is impaired in cells that lack the Ku70/80 heterodimer. 




Similarly, the activation of ATM signaling at telomeres needs to be averted. When TRF2 is deleted, most telomeres are recognized by the ATM kinase pathway, leading to DNA damage foci at telomeres (called Telomere Dysfunction Induced Foci or TIFs) that contain γ-H2AX, MDC1, 53BP1 and other DDR factors. ATM kinase activation at telomeres involves recognition of the telomere end by the Mre11/Rad50/Nbs1 (MRN complex).

In collaboration with Jack Griffith (University of North Carolina, Chapel Hill) we found that telomeres can occur in a lariat conformation, referred to as the t-loop. T-loops are formed through the strand invasion of the 3’ telomeric overhang into the duplex part of the telomere. Since the discovery of t-loops in mammals, they have been found in many other eukaryotes, including protozoa, plants, and some fungi.




Given that the telomere terminus is sequestered in the t-loop configuration we proposed that this structure would protect telomeres. Specifically, the t-loop structure would render telomeres impervious to c-NHEJ, which requires the loading of the Ku70/80 complex on free DNA ends, and would prevent the activation of the ATM kinase, which involves binding of the MRN complex to DNA ends. TRF2, the only shelterin protein required for the repression of c-NHEJ and ATM signaling, has the ability to make t-loops in vitro. We tested the TRF2/t-loop model by using super-resolution STORM imaging to detect t-loops in relaxed chromatin from cells with and without TRF2. The results demonstrated that TRF2, but not the other components of shelterin, is required for the establishment and/or maintenance of t-loops.




Generation of the 3' overhang
Relevant Publications

The protection of telomeres is in part dependent on the presence of a 3’ overhang at the telomere terminus. This overhang has to be regenerated every time telomeres are duplicated. Overhang generation is a complex process that involves multiple steps and the telomeres formed by leading- and lagging-strand DNA synthesis are processed differently as expected since their terminal structures are different immediately after DNA synthesis. The process of 3’ overhang formation is carefully controlled by shelterin. TRF2 recruits the Apollo nuclease, which is critical for an initial processing step at the leading-end telomeres. POT1b on the other hand, limits the length of the 3’ overhang by inhibiting Apollo. The Exo1 exonuclease also acts on telomere ends and excessive resection by Exo1 is also counteracted by POT1b. The latter regulation involves the interaction between POT1b and the CST (Ctc1, Stn1, Ten1) complex, which can promote a Polymerase a/primase dependent fill-in reaction at telomere ends. Mice lacking POT1b show excessive telomere shortening, especially when telomerase is limiting. Ultimately, this telomere shortening evokes phenotypes reminiscent of Dyskeratosis congenita. A different inherited telomeropathy, Coats plus, is due to mutations in either CST or POT1, illustrating the importance of the interaction between shelterin and CST in the maintenance of sufficient telomere reserve.




Repression of ATR signaling by POT1
Relevant Publications

The constitutive ssDNA at telomeres can activate the ATR kinase. The ATR kinase signaling is activated through the binding of RPA to ssDNA and the Rad17-dependent loading of 9-1-1 on the neighboring 5’ ds/ss transition. ATR is recruited by ATRIP-dependent binding to RPA and is activated when TopBP1 interacts with the 9-1-1 complex. The 3’ overhang of mammalian telomeres is of sufficient length to bind RPA and ATR activation can occur if shelterin fails to protect the telomeres. The t-loop configuration does not protect telomeres from ATR signaling because all DNA structures needed for RPA binding and TopBP1-mediated ATR activation are present at the base of the t-loop.  

            Shelterin uses POT1 to repress ATR signaling. In human shelterin, this task is delegated to the single POT1 protein, whereas mouse shelterin has two ATR repressors, POT1a and POT1b. When POT1a and POT1b are both deleted, ATR is activated at telomeres throughout the cell cycle. As expected, this activation is dependent on ATRIP and RPA. By binding to the ss telomeric DNA, POT1 blocks the accumulation of RPA at telomeres and thereby prevents ATR activation. As RPA is much more abundant than POT1 and since POT1 and RPA have the same affinity for telomeric sequences, the tethering of POT1 to the rest of shelterin is the critical aspect of its ability to exclude RPA from the ssDNA.




TRF1 promotes telomere replication and prevents formation of fragile telomeres
Relevant Publications

Deletion of TRF1 from mouse embryo fibroblasts revealed that TRF1 functions to promote efficient replication of telomeric DNA. In absence of TRF1, replication fork stalling occurs in telomeric DNA tracts and the ATR kinase pathway is activated at telomeres. In metaphase, telomeres appear as broken or decondensed, resembling the common fragile sites (CFS) observed after treatment with aphidicolin. Indeed, aphidicolin treatment also induces the fragile telomere phenotype, indicating that telomeres are similar to the CFS. Experiments with the BLM and RTEL1 helicases indicated that TRF1 cooperates with these factors to remove replication blocks from the telomeric DNA. We have proposed that TRF1 acts by removing G4 structures that can be formed in the TTAGGG repeats and might impeded replication fork progression.




Using telomeres to study the DNA damage response
Relevant Publications

Telomeres offer distinct advantages for the study of the DDR. They represent molecularly marked sites in the genome that can be converted to sites of DNA damage by manipulation of shelterin. As a result, their structure, behavior, and processing can be studied before and after the induction of DNA damage signaling using imaging, DNA analysis, ChIP, etc. An additional advantage of telomeres is that they can be manipulated to activate either the ATM kinase (deletion of TRF2) or the ATR kinase (deletion of POT1a) or both (deletion of the whole shelterin complex). Using these telomeric tools, we have gained insight into several aspects of the DNA damage response, including DNA damage signaling, DSB resection, and DSB repair. In particular, our work has provided insights into the function of 53BP1, an intensely studied DNA damage response factor. 53BP1 accumulates at sites of DNA damage and affects the type of DSB repair at these sites. We have found that one function of 53BP1 is to endow sites of DNA damage with increased mobility in the nucleus in a manner that depends on cytoplasmic microtubules and the LINC complex. In addition, we have gained insights into how 53BP1 controls the formation of ss DNA at DSBs. We identified Rif1 as a novel DDR factor that acts together with 53BP1 to prevent ssDNA formation. Our recent work on 53BP1 and Rif1 has revealed that their ability to prevent ssDNA formation is due to CST-dependent fill-in synthesis, similar to the fill-in of the telomeric overhang. These insights into the function of 53BP1 are relevant to the treatment of BRCA1-negative breast cancers with PARP1 inhibitors.










TelomeresTelomeres, the nucleoprotein complexes at the natural ends of chromosomes, protect chromosome ends and ensure their complete replication. The loss of telomere protection is the cause of the