Functional interaction between telomere protein TPP1 and telomerase

Abstract

Human chromosome end-capping and telomerase regulation require POT1 (Protection of Telomeres 1) and TPP1 proteins, which bind to the 3' ssDNA extension of human telomeres. POT1-TPP1 binding to telomeric DNA activates telomerase repeat addition processivity. We now provide evidence that this POT1-TPP1 activation requires specific interactions with telomerase, rather than it being a DNA substrate-specific effect. First, telomerase from the fish medaka, which extends the same telomeric DNA primer as human telomerase, was not activated by human POT1-TPP1. Second, mutation of a conserved glycine, Gly100 in the TEN (telomerase essential N-terminal) domain of TERT, abolished the enhancement of telomerase processivity by POT1-TPP1, in contrast to other single amino acid mutations. Chimeric human-fish telomerases that contained the human TEN domain were active but not stimulated by POT1-TPP1, showing that additional determinants of processivity lie outside the TEN domain. Finally, primers bound to mouse POT1A and human TPP1 were activated for extension by human telomerase, whereas mPOT1A-mTPP1 was most active with mouse telomerase, indicating that these mammalian telomerases have specificity for their respective TPP1 proteins. We suggest that a sequence-specific interaction between TPP1 in the TPP1-POT1-telomeric DNA complex and the G100 region of the TEN domain of TERT is necessary for high-processivity telomerase action.

Figure 1.

Human POT1–TPP1 bound to a telomeric DNA primer enhances the processivity of human but not medaka telomerase. TERT and TR were expressed in RRLs and telomerase was immunopurified. Primer a5 (TTAGGGTTAGCGTTAGGG, 50 nM) has a single base mutation that positions POT1 on the 10 nt at the primer 5′ end (Lei et al. 2005). Direct telomerase assay (1 h) in the presence of ³²P-dGTP with primer alone (−), or in the presence of 500 nM POT1 (P), 500 nM TPP1 (T), or 500 nM of each protein (PT). Lanes +2 and +4 are markers synthesized by telomerase extension of primer (GGTTAG)₃ in the presence of dGTP only (+2) or dGTP and dTTP but no dATP (+4). They consistently run slightly higher than the corresponding products in the experimental lanes because of the slight difference in base composition of the primers. The numbers 2–30 at left indicate the number of telomeric repeats synthesized.

Figure 2.

Mutations in the TEN domain of hTERT uncouple POT1–TPP1 activation from basal activity. (A) Domain structure of TERT and crystal structure of the TEN domain of Tetrahymena TERT, shown as a surface representation. Based on the assumption that conserved amino acids occupy similar positions in the human and Tetrahymena proteins, the amino acids that define the anchor site are shown in red, and other amino acids mutated in this study are also colored. The numbering system is given as Tetrahymena TERT (human TERT). (B) Telomerase activity assays for hTERT mutants complexed with hTR. The conditions are the same as in Figure 1. (C) SDS-PAGE showing the integrity and equal concentration of the ³⁵S-labeled hTERT proteins used in B, as well as the medaka (medWT) and mouse (mWT) TERTs.

Figure 3.

Quantitation of activity and processivity of hTERT mutants confirms the exceptional nature of G100V. (A) Activity of mutant telomerases relative to wild type. Error bars represent standard error, typically five independent measurements. (B) Change in activity when primer is bound to hPOT1 (P), hTPP1 (T), or both (PT), with each activity measurement normalized to the activity of the same telomerase in the absence of hPOT1 and hTPP1. The wild-type (WT) medaka telomerase shown as the last set. (C) Processivity of mutant telomerases, with standard error as above. (D) Analysis of processivity from the gel of Figure 2B, one of the data sets used for C. Medaka data from the gel of Figure 4. (F) Fraction of telomerase products extended beyond the indicated repeat number. See the Materials and Methods for details.

Figure 4.

Chimeric human–medaka TERT is active but is not stimulated by hPOT1–TPP1. The chimeric TERT has the TEN domain of hTERT (amino acids 1–187) fused to the remainder of medaka TERT (starting with Phe185), reconstituted with medaka TR. Lanes 1–4 show activity of wild-type (WT) medaka TERT–medaka TR for comparison. Primer a5 is at 50 nM; each protein is at 500 nM as in previous figures.

Figure 5.

Human but not mouse TPP1 can activate human telomerase even in a heterologous POT1–TPP1 complex. (A) Typical purity of the mouse POT1A and TPP1 determined by SDS-PAGE. (B) Binding of mouse POT1A (squares, solid line) and human POT1 (circles, dashed line) to a trace amount of radiolabeled a5 primer determined by filter binding. (C) Formation of POT1–DNA binary complexes and POT1–TPP1–DNA ternary complexes determined by an electrophoretic mobility shift assay. Radiolabeled primer T₈GGTTAGGGTTAG (position indicated by arrow), where the eight T nucleotides are necessary to reveal the subtle gel shift obtained when TPP1 (an acidic protein) joins the complex. Human (h) and mouse (m) proteins, each at 50 nM. All combinations of POT1 and TPP1 gave a stable ternary complex (indicated by an asterisk to the left of the band), with complex formation being complete in the case of human/human (h/h), half-complete in the cases of mouse/mouse (m/m) and mouse/human (m/h), and only a low extent in the case of human/mouse (h/m). (D) Telomerase activity assays with 50 nM primer a5 in the absence (−) or presence of telomere proteins (500 nM each) as indicated. (E) Quantitation of total human telomerase activity (relative to activity of primer a5 in the absence of telomere proteins) and of processivity (in units of number of telomeric repeats) for the data shown in D and independent experiments not shown. Error bars represent standard deviation; n = 3.

Figure 6.

Telomerases that contain mouse TERT have their processivity activated by mouse POT1A–TPP1. (A) Representative activity data, with 50 nM primer a5 and 500 nM each indicated protein. All telomerases were comprised of mTERT plus the RNA indicated; mTR* is full-length mouse TR, with the asterisk indicating two point mutations in the primer alignment region that improve primer binding (see Supplemental Fig. 5B). For other RNAs, numbers represent the nucleotides present in the transcript. Activity of telomerases in the two left panels was consistently low, so these data are shown with longer exposure than in the right panel. (B) Quantitation of telomerase processivity (in units of number of telomeric repeats) for the data shown in A and independent experiments not shown. hTR/mTR indicates a two-piece RNA system consisting of nucleotides 34–192 of hTR and domains CR4 and CR5 of mTR. Error bars represent standard deviation; n ranged from 3 to 6.