Telomerase structures and regulation: shedding light on the chromosome end

Abstract

During genome replication, telomerase adds repeats to the ends of chromosomes to balance the loss of telomeric DNA. The regulation of telomerase activity is of medical relevance, as it has been implicated in human diseases such as cancer, as well as in aging. Until recently, structural information on this enzyme that would facilitate its clinical manipulation had been lacking due to telomerase very low abundance in cells. Recent cryo-EM structures of both the human and Tetrahymena thermophila telomerases have provided a picture of both the shared catalytic core of telomerase and its interaction with species-specific factors that play different roles in telomerase RNP assembly and function. We discuss also progress toward an understanding of telomerase RNP biogenesis and telomere recruitment from recent studies.

Figure 1

Telomerase holoenzymes and telomerase recruitment to telomeres in (a) human, (b) T. thermophila, (c) S. cerevisiae, and (d) S. pombe (illustrations not drawn to scale). Telomerase holoenzyme and telomere protein complex schematics are shown, together with predicted secondary structures of TERs, and their lengths in nucleotides (nt) are indicated in parenthesis. TERTs are colored in light green and other holoenzyme proteins are placed and colored based on their predicted or known interaction networks. Below the telomerase holoenzyme schematics are schematics of telomere proteins (critical ones highlighted in colors) that are important for telomerase recruitment. Black arrows indicate interactions that recruit and activate telomerase at telomeres. In (b), it is possible that T. thermophila TERT interacts with Tpt1 by analogy to human and S. pombe systems, but this interaction may be replaced by a p50-TERT interaction.

Figure 2

Cryo-EM structures of (a) human and (b) T. thermophila telomerase holoenzymes in two views [19,20]. Subunits are colored as labeled. A telomeric DNA substrate is bound to template in both structures, but the template and template-paired DNA are buried in the views shown; only the single-stranded 5’ region of unpaired DNA in the T. thermophila structure has visualized density rendered in these illustrations. In (b), the depicted 4.8 Å T. thermophila telomerase structure was obtained from focused refinement of a structural core containing only TERT, TER, p50 and some of the TEB complex; p65 and the p75-p45-p19 heterotrimer complex are absent in this map (see Figure 1a for the complete holoenzyme). Only single domains of TEB subunits were resolved (Teb1C is Teb1 C-terminal domain, Teb2N is Teb2 N-terminal domain).

Figure 3

Structural analysis of the catalytic core of the human and T. thermophila telomerase structures [19,20]. (a) Domain architecture of TERT. The IFD (light green) is embedded in the RT domain (dark green). (b) Secondary structure schematics of human and T. thermophila TERs. RNA domains described in the text are labeled; in addition, T. thermophila stem 2 (S2) is indicated, which is the template 5’ flanking region labeled in Figure 2b. The domain colors shown in (a) and (b) are used for the subsequent panels, which look down into the active site. (c and d) Catalytic cores of human and T. thermophila telomerase, respectively. (e) Close-up view of TRAP in the IFD, highlighted in space-filling representation. (f and g) Schematic representation of the architectures of the catalytic core in human and T. thermophila telomerase, respectively. The positioning of human TPP1 and POT1 is only hypothetical. The interactions observed in T. thermophila telomerase holoenzyme between TERT, p50 and TEB may or may not be paralleled by TERT, TPP1 and POT1 in human telomerase-telomere complexes.

Figure 4

Structures of protein domains and sub-complexes with important roles in telomerase recruitment to telomeres. Structures recently determined are shown in colors. The shaded wedges indicate interactions. (a) Domain architectures of human POT1, TPP1, TIN2 and TRF2. OB indicates OB-folds. POT1 has three OB-fold domains and a Holliday junction resolvase-like (HJRL) domain embedded within the third OB-fold. TPP1 has an OB-fold domain, a POT1-binding motif (PBM), and a TIN2-binding motif (TBM). TIN2 has a TRFH domain, a TRFH-binding motif (TBM) and a dyskeratosis congenita (DC) disease mutation hotspot region. TRF2 has a basic domain, a TRFH domain, a RAP1-binding motif (RBM), a TIN2-binding motif (TBM) and a Myb DNA-binding domain. (b) Structure of the third OB-fold (OB3) and HJRL domains of POT1 in complex with the PBM of TPP1 (PDB 5H65) [56,57]. (c) Structure of the TRFH domain of TIN2 in complex with the TBMs of TPP1 and TRF2 (PDB 5XYF) [58]. (d) Domain architecture of S. cerevisiae Ku70, Ku80 and Sir4. Each of Ku70 and Ku80 contains a von Willebrand factor type A domain (vWA), a β-barrel domain (BBD) and a C-terminal α-helical domain (CTD). Sir4 has a Ku-binding motif (KBM), a Sir2-interacting domain (SID), a partitioning and anchoring domain (PAD), and a coiled-coil (CC) domain. (e) Domain architecture of K. lactis Est1 and Cdc13. Est1 has a tetratricopeptide repeat domain (TPR), a helical hairpin domain (HHD), and an insertion motif (IM). Cdc13 has three OB-fold domains, OB1, OB2 and OB4, an Est1-binding motif (EBM), a recruitment domain (RD), and a DNA-binding domain (DBD). (f) Structure of S. cerevisiae Ku70/Ku80 and the Ku binding site of TLC1 (PDB 5Y58) [25]. (g) Structure of S. cerevisiae Ku80 vWA domain with Sir4 KBM (PDB 5Y59) [25]. (h) Structure of K. lactis Est1 in complex with EBM of Cdc13 (PDB 5Y5A) [25].