PinX1 localizes to telomeres and stabilizes TRF1 at mitosis

Abstract

Human telomeres are DNA-protein complexes that cap and protect the ends of chromosomes. The protein PinX1 associates with telomeres through an interaction with the resident double-stranded telomere-binding protein TRF1. PinX1 also binds to and inhibits telomerase, the enzyme responsible for complete replication of telomeric DNA. We now report that endogenous PinX1 associates with telomeres primarily at mitosis. Moreover, knockdown of PinX1 caused delayed mitotic entry and reduced the accumulation of TRF1 on telomeres during this stage of the cell cycle. Taking these findings together, we suggest that one function of PinX1 is to stabilize TRF1 during mitosis, perhaps to promote transition into M phase of the cell cycle.

Fig 1

PinX1 localizes at telomeres during M phase. (A) Levels of endogenous PinX1, as assessed by immunoblot assay with an anti-PinX1 antibody, in cells untreated or treated with two different PinX1 siRNAs. Data are representative of two replicate experiments. (B to D) HeLa cells were synchronized by double thymidine block, released, and collected at the four indicated time points. Collected cells were split into three portions that were stained with propidium iodide and subjected to FACS for cell cycle analysis (B), immunoblotted to detect endogenous PinX1, phospho-histone H3 (P-HH3) as a marker of mitosis, or tubulin as a loading control (as additional controls, PinX1 and tubulin levels were also analyzed in scrambled-siRNA-transfected cells, mock-transfected cells, or untransfected cells synchronized by double thymidine block) (C), or used for ChIP analysis with an anti-PinX1 antibody or an anti-TRF1 antibody or IgG as controls, followed by Southern hybridization with a telomere or control Alu probe (5 μg genomic DNA served as a hybridization control) (D). Data are representative of three replicate experiments. (E) Means ± SD of the normalized hybridization intensities of telomeric DNA coimmunoprecipitated with endogenous PinX1 from three independent ChIP experiments at the indicated cell cycle phases.

Fig 2

Knockdown of PinX1 delays entry into mitosis. HeLa cells transfected with a scrambled control siRNA or PinX1 siRNA were synchronized by double thymidine block, released, and collected at the indicated six time points. (A and B) Lysates from cells at each time point and asynchronous cells were either immunoblotted to detect endogenous PinX1, phosphorylated histone H3 (P-HH3) as a marker of mitosis, or tubulin as a loading control (A) or stained with propidium iodide and subjected to FACS for cell cycle analysis (black trace, scrambled-siRNA-treated control cells; red trace, PinX1 siRNA-treated cells) (B). Data are representative of five replicate experiments. (C and D) Mean length of time ± SD to enter (C) or to complete (D) mitosis beginning 8 h after release from a double thymidine block for 23 HeLa cells stably expressing GFP-histone H2B and transfected (as verified by BLOCK-iT Alexa Fluor Red oligonucleotide fluorescence) with either a scrambled control siRNA (●) or a PinX1 siRNA (■). **, P < 0.01; N.S., not significant. (E) HeLa cells transfected with a scrambled control siRNA or a PinX1 siRNA were synchronized by double thymidine block, released, collected at the time points indicated above the gels, and immunoblotted to detect endogenous PinX1, phosphorylated histone H3 (P-HH3), and cleaved caspase 3. Data are representative of two replicate experiments. A, asynchronous cells.

Fig 3

TIN2 inhibits PinX1 binding to TRF1. (A) Association of rabbit reticulocyte lysate (RRL)-generated ³⁵S-labeled recombinant TIN2 with bacterially generated recombinant GST-TRF1 upon coincubation with increasing concentrations of recombinant MBP-PinX1. GST and MBP served as controls. The relative amounts of TIN2 normalized to the amount detected with GST-TRF1 are shown below the gel. Data are representative of two replicate experiments. (B) Association of RRL-generated ³⁵S-labeled recombinant PinX1 with bacterially generated recombinant GST-TRF1 upon coincubation with increasing concentrations of recombinant MBP-TIN2. GST and MBP served as controls. The relative amounts of PinX1 normalized to the amount detected with GST-TRF1 are shown below the gel. Data are representative of two replicate experiments. (C) Levels of Myc-TRF1 detected by immunoprecipitation (IP) of FLAG-PinX1 followed by immunoblot assay with an anti-Myc antibody in 293T cells expressing HA-TIN2. Tubulin served as a loading control. The relative amounts of Myc-TRF1 coimmunoprecipitated with FLAG-PinX1 normalized to tubulin are shown below the gel. Data are representative of two replicate experiments. (D) Levels of Myc-TRF1 detected by immunoprecipitation of HA-TIN2 followed by immunoblot assay with an anti-Myc antibody in 293T cells expressing FLAG-PinX1. Tubulin served as a loading control. The relative amounts of Myc-TRF1 coimmunoprecipitating with HA-TIN2 normalized to tubulin are shown below the gel. Data are representative of two replicate experiments.

Fig 4

Knockdown of PinX1 decreases TRF1 levels at M phase. (A) Myc-TRF1 levels in HeLa cells treated with PinX1 or scramble siRNA, as assessed by immunoblot assay. Tubulin served as a loading control. The relative amounts of Myc-TRF1 normalized to tubulin are shown below the gel. Data are representative of two replicate experiments. (B) HeLa cells treated with a scrambled control siRNA or PinX1 siRNA and either left untreated (interphase cells) or treated with nocodazole and subjected to mitotic shake-off (mitotic cells) were immunoblotted to detect phospho-histone H3 (P-HH3) as a mitotic marker, PinX1, TRF1, or tubulin as a loading control. The relative amounts of TRF1 normalized to tubulin are shown below the gel. Data are representative of two replicate experiments. (C) Amount of Myc-TRF detected by immunoprecipitation (IP) of HA-Fbx4 followed by immunoblot assay with an anti-Myc antibody in 293T cells with increasing amounts of FLAG-PinX1. Tubulin served as a loading control. The relative amounts of Myc-TRF1 coimmunoprecipitated with HA-Fbx4 normalized to tubulin are shown below the gel. Data are representative of three replicate experiments.

Fig 5

Knockdown of PinX1 delays the reloading of TRF1 onto telomeres during mitosis. HeLa cells transfected with a scrambled control siRNA (A) or a PinX1 siRNA (B) were synchronized by double thymidine block, released, and collected at the indicated time points after release from the block. Collected cells were split into two portions for ChIP and immunoblot analysis. ChIP analysis was performed with an anti-TRF1 antibody or IgG as a control, followed by Southern hybridization with a telomere or control Alu probe (genomic DNA served as a hybridization control). Immunoblot analysis was performed to detect phosphorylated histone H3 (P-HH3) as a marker of mitosis or tubulin as a loading control. Data are representative of five (A) and three (B) replicate experiments. (C and D) Normalized hybridization intensities of telomeric DNA coimmunoprecipitating with endogenous PinX1 at the indicated time points in scrambled (C) or PinX1 (D) siRNA-treated cells. (E) Asynchronous (interphase) HeLa cells or HeLa cells treated with nocodazole and subjected to mitotic shake-off to enrich metaphase (M) cells were subjected to ChIP analysis and immunoblotted for phosphorylated histone H3 or tubulin as described above. Data are representative of two replicate experiments.