The E3 ubiquitin ligase Rnf8 stabilizes Tpp1 to promote telomere end protection

Abstract

The mammalian shelterin component TPP1 has essential roles in telomere maintenance and, together with POT1, is required for the repression of DNA damage signaling at telomeres. Here we show that in Mus musculus, the E3 ubiquitin ligase Rnf8 localizes to uncapped telomeres and promotes the accumulation of DNA damage proteins 53Bp1 and γ-H2ax. In the absence of Rnf8, Tpp1 is unstable, resulting in telomere shortening and chromosome fusions through the alternative nonhomologous end-joining (A-NHEJ) repair pathway. The Rnf8 RING-finger domain is essential for Tpp1 stability and retention at telomeres. Rnf8 physically interacts with Tpp1 to generate Ubc13-dependent Lys63 polyubiquitin chains that stabilize Tpp1 at telomeres. The conserved Tpp1 residue Lys233 is important for Rnf8-mediated Tpp1 ubiquitylation and localization to telomeres. Thus, Tpp1 is a newly identified substrate for Rnf8, indicating a previously unrecognized role for Rnf8 in telomere end protection.

Figure 1. RNF8 is required for the accumulation of DNA damage response factors to dysfunctional telomeres

(a) 53BP1 and (b) γ-H2AX positive TIFs in Rnf8^+/+ and Rnf8^−/− MEFs after depletion of Trf2 or mTPP1. Cells fixed after 72 hours stained with anti-53BP1 antibody (green) or anti γ-H2AX antibody (green), with Tam-OO-(CCCTAA)₄ telomere peptide nucleic acid (red) and 4,6-diamidino-2-phenylindole (DAPI; blue). (c) Quantification of 53BP1 and γ-H2AX positive TIFs. A minimum of 100 cells were examined and cells with >4 TIFs were scored as TIF positive. Mean values were derived from at least three experiments. Error bars: s.d. Rnf8 positive foci in Rnf8^−/− MEFs reconstituted with wild-type mRNF8 and indicated mRNF8 point mutants, followed by (d) depletion of Trf2 or (e) mTPP1. Cells were fixed, stained with anti-FLAG antibody (green), Tam-OO-(CCCTAA)₄ telomere peptide nucleic acid (red) and DAPI (blue). (f) Quantification of mRNF8-positive foci observed in (d and e). A minimum of 100 cells were examined and cells with >4 TIFs were scored as TIF positive. Mean values were derived from at least three experiments. Error bars: s.d.

Figure 2. RNF8 is required to protect telomere ends from chromosome aberrations

(a) Telomere-FISH analysis on metaphase chromosome spreads of immortalized Rnf8^+/+ and Rnf8^−/− MEFs using Tam-OO-(CCCTAA)₄ telomere peptide nucleic acid (red) and DAPI (blue). A minimum of 30 metaphases were analyzed per genotype. Red arrows-ring chromosomes; orange arrows: radial chromosomes; yellow arrows: p-p chromosome fusions without telomere at the site of fusions; green arrows: telomere signal-free chromosome ends. (b) Quantification of chromosome aberrations in Rnf8^+/+ and Rnf8^−/− metaphases. A minimum of 30 metaphases were examined. Mean values were derived from at least three experiments. Error bars: s.d. (c) SV40LT immortalized Rnf8^+/+ and Rnf8^−/− MEFs were treated with control vector or mTPP1^ΔRD for 120 h, metaphases were prepared and telomere aberrations visualized by Tam-OO-(CCCTAA)₄ telomere peptide nucleic acid (red) and DAPI (blue). Arrows: fused chromosomes. (d) Quantification of telomere fusion frequencies in Rnf8^+/+ and Rnf8^−/− MEFs after expressing mTPP1^ΔRD. A minimum of 1200 chromosomes were analyzed and mean values derived from at least three experiments are presented. Error bars: s.d. *P = 0.003, two-tailed Student's t-test. (e) SV40LT immortalized Rnf8^+/+ and Rnf8^−/− MEFs were treated with control vector or shTrf2 for 120 h, metaphases prepared and telomere aberrations visualized by Tam-OO-(CCCTAA)₄ telomere peptide nucleic acid (red) and DAPI (blue). Arrows: fused chromosomes in Rnf8^−/− MEFs treated with mTPP1^ΔRD. The numerous fused chromosomes seen in Rnf8^+/+ MEFs treated with shTrf2 are not labeled. (f) Quantification of telomere fusion frequencies in Rnf8^+/+ and Rnf8^−/− MEFs after Trf2 depletion. A minimum of 1200 chromosomes were analyzed and mean values derived from at least three experiments presented. Error bars: s.d. *P = 0.01, **P = 0.003, two-tailed Student's t-test.

Figure 3. RNF8 is required for the accumulation of the TPP1-POT1a/b complex at telomeres

(a) Rnf8^+/+ and Rnf8^−/− cell lysates were subjected to SDS-PAGE and Western blotting with the indicated antibodies. The specificity of the mouse TPP1 antibody is demonstrated by shRNA against mTPP1, which efficiently depleted both endogenous and exogenous mTPP1. γ-tubulin was used a loading control. (LE: long exposure, SE: short exposure). Quantification of TPP1 expression levels: lane 1–1.0, lane 2–0.41, lane 3–0.34, lane 4–0.20. Quantification of mTRF2 levels: lane 1–1.0, lane 2–1.0, lane 3–0.86, lane 4–1.2. b) HeLa cells transfected with two independent siRNAs against RNF8 were harvested after 48 hours, subjected to SDS-PAGE and imunoblotted with anti-hRNF8, anti-hTPP1 and anti-hTRF2 antibodies. γ-tubulin was used as a loading control. Quantification of protein levels are as indicated. (c) Rnf8^+/+ and Rnf8^−/− cells grown on coverslips were stained with anti-mTPP1 antibody (green), Tam-OO-(CCCTAA)₄ telomere peptide nucleic acid (red) and DAPI (blue). The specificity of the mouse TPP1 antibody for immunofluorscence is demonstrated in the upper panels by shRNA against mTPP1 (d). Quantification of mTPP1 co-localized with Tam-OO-(CCCTAA)₄ telomere peptide nucleic acid. A minimum of 100 cells were examined mean values were derived from three experiments. Error bars: s.d. **P = 0.001, two-tailed Student's t-test. (e) Rnf8^−/− MEFs reconstituted with wild-type mRNF8 and the mRNF8^C406S RING mutant were stained with anti-mTPP1 antibody (green), Tam-OO-(CCCTAA)₄ telomere peptide nucleic acid (red) and DAPI (blue).

Figure 4. RNF8 is required for preventing TPP1 from proteasome-mediated degradation

(a) Rnf8^+/+ and Rnf8^−/− MEFs were incubated in the presence or absence of 12.5μ-M of MG132 and lysates subjected to immunoblotting with anti-mTPP1 antibody. (b) Lysates isolated from Rnf8^+/+ and Rnf8^−/− MEFs stably expressing HA-mTPP1 in the presence or absence of 12.5μ-M MG132 were subjected to immunoblotting with anti-HA antibody. γ-Tubulin used as a loading control. (c) Rnf8^+/+ and Rnf8^−/− MEFs transfected individually with the indicated cDNA constructs in the presence or absence of 12.5μ-M of MG132 were subjected to immunoblotting with anti-FLAG and anti-HA and (d) anti-Myc antibodies. γ-Tubulin served as the loading control. (e) Cells grown on coverslips in the presence of 12.5 μm MG132 were fixed and stained with anti-mTPP1 (green), Tam-OO-(CCCTAA)₄ telomere peptide nucleic acid (red) and DAPI (blue). (f) Rnf8^+/+ and Rnf8^−/− MEFs transfected with HA-mTPP1 and incubated in the presence or absence of 12.5μ-M of MG132 were subjected to immunofluorescence microscopy with anti-HA antibody (green), telomere-PNA FISH (red) and DAPI (blue).

Figure 5. RNF8 physically interacts with and ubiquitylates TPP1

(a) 293T cells transfected with the indicated plasmids were immunoprecipitated with anti-HA antibody, subjected to SDS PAGE and western blotting with the indicated antibody. Input represent 5% of the total cell lysate used for the immunoprecipitations. (b) The ubiquitylation assay was performed with purified mRNF8^WT or the mRNF8^C406S RING mutant proteins, together with purified HA-hTPP1 protein and ubiquitin-activating enzyme (E1), MMS2 and Ubc13, RNF8, ubiquitin, ubiquitin aldehyde and ATP. Reaction mix subjected to immunoblotting to detect ubiquitylated hTPP1 with the anti-hTPP1 antibody. Anti-hRNF8 antibody used to show equal amount of hRNF8 used in the reactions. The HA-hTPP1-Ub species migrates at 250kD. (c) Stably infected mTPP1 in HeLa-UB cell line stably expressing treated with 2μg/ml biotin were harvested and incubated with strepavidin-agarose slurry. Washed beads eluted subjected to SDS-PAGE and Western blotting with the indicated antibodies. Stably infected mTPP1 cells co-infected with shTpp1 was used as a control to show the specificity of mTPP1 ubiquitylation. γ-tubulin was used as the loading control. Bracket indicates smear signals indicative of polyubiquitylation. (d) Lysates obtained from HeLa cells transfected with the indicated constructs or siRNF8 were incubated with nickel-NTA-agarose beads. Washed beads eluted subjected to SDS-PAGE and Western blotting with the indicated antibodies. Anti-HA and anti-ubiquitin antibodies were used to detect the expression of HA-mTPP1 and ubiquitin. γ-Tubulin was used as a loading control. (e) In vitro ubiquitylation assay performed as in (b) with either wild-type His-tagged ubiquitin or His-tagged ubiquitin mutants in which all lysines were mutated to arginines except for the one indicated. Reaction mix was subjected to immunoblotting with TPP1 (top panel) or anti-HA antibody (lower panel). The HA-hTPP1-Ub species migrates at 250kD.

Figure 6. Lysine 233 is required for TPP1 ubiquitylation, telomere localization and genome stability

(a) Schematic representation of the conserved lysine residues in hTPP1 across the species (lysine residue mutated are highlighted in yellow). (b) Lysine K233 is required for hTPP1 ubiquitylation. In vitro ubiquitylation assay in the presence and absence of purified wild-type mRNF8 with either wild type HA-hTPP1 or indicated lysine to arginine HA-hTPP1 mutants. Reaction mix was subjected to immunoblotting with anti-HA antibody. The HA-hTPP1-Ub species migrates at 250kD. Bottom panel is a short exposure to show equal amounts of TPP1 proteins used in the assay. (c) Telomere-PNA FISH showing the localization of exogenous hTPP1 lysine mutants in Rnf8^+/+ MEFs. Cells stably infected with FLAG-tagged wild-type or indicated hTPP1 mutants were stained with anti- FLAG antibody (green), Tam OO- (CCCTAA)₄ telomere peptide nucleic acid (red) and DAPI (blue). (d) Quantification of localization of wild-type and mutant hTPP1 on telomeres. A minimum of 200 cells were examined mean values were derived from three experiments. Error bars: s.d. *P = 0.001, **P = 0.0006, two-tailed Student's t-test.