Eukaryotic genome integrity and stability rely on the presence of telomeres, specialized protein-DNA complexes that compose the natural termini of linear chromosomes. Telomeres form caps at the ends of chromosomes that protect them from events that would be catastrophic for the genome. Telomeric DNA consists of tandem repeats of a short G-rich sequence oriented 5' to 3' toward the chromosome terminus, with the G-rich strand extending beyond its complement to form a 3' G-overhang (Figure 1).
Loss of telomeric integrity leads to chromosomal instability. This leads to various chromosomal rearrangements, including formation of dicentric chromosomes and gene amplification, which may lead to oncogene activation and cancer. In most eukaryotes, telomere length is maintained by telomerase, a specialized reverse transcriptase, which adds telomeric DNA to the 3' ends of chromosomes to ensure complete genome replication (Figure 1). Remarkably, telomerase is also strongly up-regulated in most cancer cells and has been studied as a plausible anticancer target.
A six-protein complex, called shelterin, protects the telomeres of human chromosomes (Figure 1). Proteins TRF1 and TRF2 directly bind double-stranded telomeric DNA, POT1 directly binds the single-stranded 3' extension at the chromosome end, and these are bridged through protein-protein interactions involving TIN2 and TPP1. The sixth protein, RAP1, binds mostly to TRF2. Two functions have been proposed for this complex: protection of the natural chromosome end from being mistaken for a broken end and being subjected to DNA repair, and negative regulation of telomerase by sequestration of its telomeric DNA substrate.
Although much information has been accumulated about the protein components of human telomeres, the molecular mechanisms by which shelterin mediates maintenance and protection of human telomeres remain poorly understood. Our long-term goal is to paint a detailed picture of how telomeres protect chromosome ends and mediate their replication by telomerase.
Single-Stranded Telomere-Binding Protein Complex POT1-TPP1 and Telomerase Regulation
Proteins that bind to the single-stranded telomeric G-overhang were first identified in ciliated protozoa. The telomere endbinding protein (TEBP) from Oxytricha nova is composed of two subunits, TEBP? and TEBP? (Figure 2A). POT1 proteins were identified from many organisms from fission yeast to plants and mammals based on their limited sequence homology with TEBP?. POT1 protein binds the single-stranded telomeric overhangs and suppresses unwanted repair activities (Figure 2B). It provides the most widespread solution to chromosome end protection in eukaryotes.
Although POT1 is evolutionarily conserved, bioinformatics approaches failed to identify a TEBP? ortholog in any organism other than protozoa. Whether higher eukaryotes have a TEBP? homolog had been a mystery. TPP1 was identified as a shelterin protein that simultaneously interacts with both POT1 and TIN2 (Figure 1). To explore the additional functions of TPP1 at the telomere, we determined the crystal structure of a portion of TPP1 (Figure 2C). This structure reveals an oligonucleotide/oligosaccharide-binding (OB) fold that is structurally most similar to the ? subunit of O. nova TEBP (Figure 2D). Our biochemical studies also demonstrated that, although TPP1 itself does not bind single-stranded DNA (ssDNA), it enhances the POT1-ssDNA interaction, closely resembling the properties of TEBP?. Thus, our data showed that TPP1 is the missing human homolog of the O. nova TEBP ? subunit and that capping of telomeres by a TEBP?-? dimer is more conserved evolutionary than had been expected.
Forming telomeres of optimal length during the developmental stages when telomerase is expressed is crucial for long-term survival. But how is this achieved? We demonstrated that TPP1 plays a central role in controlling this process. Our data showed that POT1-TPP1 can both positively and negatively regulate telomerase activity. When POT1-TPP1 binds to the telomeric 3' end, it inhibits telomerase activity. Strikingly, when POT1-TPP1 binds to a more upstream register, leaving a telomerase-extendable 3' tail, it not only improves telomerase activity but also greatly increases telomerase processivity (Figure 2E). This stimulatory effect requires the POT1-TPP1 interaction and specifically the OB fold of TPP1, which, we showed, physically interacts with TERT (the catalytic component of telomerase) and thus mediates direct communication between shelterin and telomerase. This work is a significant step toward understanding the mechanism of telomere elongation by telomerase.
Double-Stranded Telomere-Binding Proteins TRF1 and TRF2 Mediate Recruitment of Telomere-Associated Factors
Shelterin is not a static structural component of the telomere. Instead, shelterin is a protein complex that acts with many associated proteins (such as DNA repair factors) to change the structure of the telomere, thereby protecting and regulating chromosome ends. Shelterin has a startling number of interacting partners that function in DNA processing and regulation. It is believed that these protein factors, in concert with shelterin, may suppress the inappropriate fusion or recombination of telomeres. Notably, most of these shelterin-mediated interactions involve TRF1 and/or TRF2.
Because of their abilities to interact with multiple proteins, TRF1 and TRF2 are by definition hubs of the protein-protein interactions at the telomeres. How do TRF1 and TRF2 mediate these interactions? Can TRF1 and TRF2 interact with multiple proteins simultaneously? To answer these questions, we have used structural and biochemical approaches to dissect the interactions of TRF1 and TRF2 with their shared binding partner, TIN2, and other shelterin accessory factors. We discovered that TRF1 recognizes a short peptide of TIN2 through its TRFH (TRF homology) domain (Figure 3A). This surface of TRF2 does not, however, bind TIN2, and TIN2 binds to a region of TRF2 outside TRF2TRFH. Instead, TRF2TRFH binds a short peptide of a shelterin accessory factor, Apollo, which is a DNA exonuclease (Figure 3B). Apollo does not interact with the TRF1TRFH. Conversely, TRF1TRFH, but not TRF2TRFH, interacts with another shelterin-associated factor, PinX1, a telomerase inhibitor.
Our data have indicated that binding to the TRFH docking site involves the sequence F/Y-x-L-x-P (F-x-L-x-P for TRF1TRFH and Y-x-L-x-P for TRF2TRFH) in the shelterin-associated factors, which contacts the same surface of theTRFH domains (Figure 3C and 3D). Thus, our work established that the TRFH domains of TRF1 and TRF2 are a novel family of protein modules that recognize short peptide motifs and function as protein-docking sites to recruit different factors to telomeres by specifically interacting with either TRF1 or TRF2.
Our discovery of the TRFH-binding consensus sequence provides a framework for the identification and understanding of other TRFH-mediated interactions. Protein sequence database searches identified many instances of telomere-associated factors containing the F/Y-x-L-x-P motif. Currently, we are investigating the biological significance of the potential TRF1/TRF2-binding proteins at telomeres.
Post-translational Modification of a Telomere Length Regulator
TRF1 is a key regulator of telomere length homeostasis. TRF1 binds along telomeres and functions as a measuring device to assess telomere length. Therefore, the number of TRF1 molecules at telomeres must be tightly controlled. Recent studies have suggested that this process involves multiple post-translational modification events of TRF1. TRF1 is ADP-ribosylated by tankyrase-1, a telomeric poly(ADP-ribose) polymerase. ADP-ribosylation of TRF1 inhibits TRF1 binding to telomeres, and overexpression of tankyrase-1 in cells releases TRF1 from telomeres and induces telomere elongation. TRF1 is also ubiquitinated and degraded through the proteasome pathway, a process mediated by FBX4, a novel F-box protein. F-box proteins function as substrate-specific adaptors for the SCF ubiquitin ligase.
As with other F-box proteins, association of FBX4 with TRF1 involves sequences C-terminal to the F-box motif (FBX4CTD). However, unlike most characterized F-box proteins, FBX4 lacks previously identified protein interaction domains outside of the F-box motif. We have determined the crystal structure of the TRF1TRFH-FBX4CTD complex (Figure 4A). Notably, FBX4CTD contains one structural domain similar to small GTPase proteins (Figure 4B). However, sequence analysis indicates that FBX4 does not have the highly conserved residues crucial for nucleotide binding. Follow-up biochemical experiments confirmed that FBX4 indeed does not bind GTP or GDP. Thus, the GTPase domain of FBX4 functions as a substrate recruitment domain rather than as a GTP-hydrolyzing enzymatic domain.
Comparison of the structure of TRF1TRFH-FBX4CTD with that of TRF1TRFH-TIN2 revealed that FBX4 and TIN2 cannot bind to TRF1 simultaneously (Figure 4C). We found that TIN2 binds more strongly than FBX4 to TRF1 and that addition of TIN2 blocks the ubiquitination of TRF1 by SCFFBX4. This is consistent with previous findings that telomere-associated (i.e., TIN2-bound) TRF1 is immune to ubiquitin-mediated degradation. Once TRF1 is ADP-ribosylated by tankyrase-1 and released from the telomeres, it no longer binds to TIN2 and thus is recognized by FBX4 for ubiquitination and degradation. Thus, FBX4 controls the pool of TRF1 that is not associated with telomeres.
In many instances, the association of targets with F-box proteins relies on an upstream signaling event, such as phosphorylation. Our work reveals a novel upstream signaling process—ADP-ribosylation. The unique feature of this regulation is that FBX4 does not directly recognize the ADP-ribosylation site on TRF1. Instead, ADP-ribosylation releases TRF1 from the telomeres so that FBX4 can recognize TRF1 for ubiquitination and degradation.
These studies are also supported by the National Institutes of Health, the Sidney Kimmel Foundation for Cancer Research, and the American Cancer Society.
As of May 30, 2012