A role for Tau protein in maintaining ribosomal DNA stability and cytidine deaminase-deficient cell survival

Abstract

Cells from Bloom's syndrome patients display genome instability due to a defective BLM and the downregulation of cytidine deaminase. Here, we use a genome-wide RNAi-synthetic lethal screen and transcriptomic profiling to identify genes enabling BLM-deficient and/or cytidine deaminase-deficient cells to tolerate constitutive DNA damage and replication stress. We found a synthetic lethal interaction between cytidine deaminase and microtubule-associated protein Tau deficiencies. Tau is overexpressed in cytidine deaminase-deficient cells, and its depletion worsens genome instability, compromising cell survival. Tau is recruited, along with upstream-binding factor, to ribosomal DNA loci. Tau downregulation decreases upstream binding factor recruitment, ribosomal RNA synthesis, ribonucleotide levels, and affects ribosomal DNA stability, leading to the formation of a new subclass of human ribosomal ultrafine anaphase bridges. We describe here Tau functions in maintaining survival of cytidine deaminase-deficient cells, and ribosomal DNA transcription and stability. Moreover, our findings for cancer tissues presenting concomitant cytidine deaminase underexpression and Tau upregulation open up new possibilities for anti-cancer treatment.Cytidine deaminase (CDA) deficiency leads to genome instability. Here the authors find a synthetic lethal interaction between CDA and the microtubule-associated protein Tau deficiencies, and report that Tau depletion affects rRNA synthesis, ribonucleotide pool balance, and rDNA stability.

Fig. 1

RNAi-based synthetic interaction screen. a Schematic representation of the genome-wide shRNA-screening procedure. Briefly, the screening involved four steps: (1) transduction of BS-Ctrl_(BLM) and BS-BLM cells with a library of pooled lentiviral human shRNAs; (2) isolation of genomic DNA from cell populations 40 h or 20 days post-infection (p.i.); (3) PCR amplification of isolated genomic DNA; (4) quantification of shRNA populations by barcode sequencing. b Identification of pathway enrichment for the list of 959 synthetic lethal genes, with the DAVID database. The x axis corresponds to –log₂(P). GO Gene Ontology. c Venn diagram showing mRNAs identified as differentially expressed in BS-Ctrl_(BLM) and BS-BLM cells, and the number of mRNAs concomitantly synthetic lethal

Fig. 2

Tau silencing reduces the survival of CDA-deficient cells. a MAPT mRNA levels determined by RT-qPCR in BS-Ctrl_(BLM) and BS-BLM cells. b, c BLM and Tau levels determined by western blotting in BS-Ctrl_(BLM) and BS-BLM cells using BLM and either Tau-5 (b) or Tau-1 (c) antibodies. d, e Tau protein quantification for Tau-5 (d) and Tau-1 (e) immunoreactivity. The β-actin signal was used as the control. The results are normalized against those for BS-BLM, which were set to 1. f–k Cells were first transfected with either non-targeting or Tau siRNA. After 24 h, cells were again transfected with the indicated siRNAs. Two days after the second round of transfection, the cells were collected for western blotting f, i, or were plated, in a serial dilution series, in 12-well plates. Ten to 12 days later, colonies were fixed and stained with crystal violet. Non-targeting siRNA was used as a control, with values set to 1 h, k. g, j Tau protein quantification after Tau-targeting siRNA depletion. l MAPT mRNA and m Tau protein levels determined in HeLa-Ctrl_(Tau) and HeLa-shTau cells. n, o HeLa-Ctrl_(Tau) and HeLa-shTau cells were first transfected with the indicated siRNAs. After 24 h, cells were again transfected with the same siRNAs. Two days after the second round of transfection, cells were collected for western blotting (n), or were plated, in a serial dilution series, in 12-well plates. Seven days later, colonies were fixed and stained with crystal violet. Non-targeting siRNA was used as a control, with the values obtained set to 1 (o). For qPCR, the mean value for 3 independent experiments is represented as a fold-change. B2M, β-actin, and TBP were used as housekeeping genes for qPCR normalization. For western blots, β-actin was used as a loading control. Each data bar is the mean of at least three independent experiments performed in triplicate. Error bars represent ± SEM. The significance of differences was assessed in two-tailed paired Student’s t-tests. ***P < 0.0005, **P < 0.005, *P < 0.05, NS not significant

Fig. 3

Tau interacts with rDNA and regulates rDNA transcription. a Immunofluorescence microscopy showing γ-H2AX (red) and 53BP1 (green) labeling. Merged images in the right panel. Scale bar: 5 µm. b γ-H2AX and Tau-1 protein levels assessed by immunoblotting. GAPDH was used as a loading control. c, d Cells were immunostained with antibodies against γ-H2AX and 53BP1. At least 400 cells were acquired, and foci counted. The number of cells (n) is indicated. e Mean number of PICH-coated UFBs per anaphase (at least 200 anaphase cells per condition). The significance of differences was assessed in two-tailed unpaired Student’s t-tests. f–i, k Cells were exposed, for 72 h, to aphidicolin (f), camptothecin (g), hydroxyurea (h), and CX-5461 (i, k). Each data point is the mean of at least three independent experiments in triplicate. j BLM, Tau-5 and CDA protein levels. β-actin was used as a loading control. l Immunofluorescence microscopy showing UBTF (red) and Tau-1 (green) co-localization in BS cells. Merged images (yellow) in the bottom panel. Scale bar: 5 µm. m Schematic representation of a human rDNA repeat (upper panel). Chromatin immunoprecipitation (ChIP) with Tau-1, UBTF and IgG antibodies in BS cells (lower panel). DNA was quantified by qPCR with primer pairs covering the rDNA repeat. Their approximate positions relative to the transcription start site (TSS) are indicated on the x axis. Data are the means from three independent experiments. n, o Endogenous pre-rRNA (45  S) levels were monitored by qPCR. B2M, β-actin, and TBP were used as housekeeping genes for normalization. The significance of differences was assessed in two-tailed paired Student’s t-tests. p, q ChIP with Tau-1 (p), UBTF (q), and IgG (p, q) antibodies. Results are obtained as described in (m). r, s Measurements of pools of NMPs and NTPs. Values of peak surface areas ratios between the endogenous nucleotide and its internal standard are shown. The data are the means of 9 independent measurements corresponding to three independent experiments performed in triplicate. The significance of differences was assessed in Mann–Whitney tests. For IF, DNA was visualized by DAPI staining (blue or white). Error bars represent ± SEM from three independent experiments. ***P < 0.0005, **P < 0.005, *P < 0.05

Fig. 4

Tau loss alters genetic integrity of rDNA and CDA^−/− cells. a, b Representative immunofluorescence (IF) z‐projection images showing paired Tau foci (green) linked by PICH-positive UFBs in BS (a) and HeLa (b) anaphase cells. Scale bar: 5 µm. DNA was visualized by DAPI staining (white). Centromeres were stained with CREST serum (blue) and UFBs were stained with PICH antibody (red). In the enlarged images, Tau foci at the extremities of UFBs are indicated by yellow arrows. c Mean number of total and Tau-positive UFBs per anaphase cell. d, f Mean number of R-UFBs per anaphase cell in untreated cell lines. e Relative number of R-UFBs per anaphase cell in BS-Ctrl_(BLM) and BS-BLM cells left untreated or treated with 100 µM tetrahydrouridine (THU) for 2 × 48 h or 1 mM of deoxycytidine (dC) for 10 h. g Mean number of R-UFBs per anaphase cell in BS-Ctrl_(BLM) and BS-BLM cells left untreated or treated with 0.4 µM aphidicolin (APH) or 1 µM ICRF-159 for 24 h. The results were normalized against those for control conditions (mock), which were set to 1. h HeLa-Ctrl_(Tau) and HeLa-shTau cells were first transfected with either non-targeting or CDA siRNA. After 48 h, cells were again transfected with the indicated siRNAs. Three days after the second round of transfection, cells were collected for western blotting, with GAPDH as the loading control (h), or were immunostained with antibody against γ-H2AX or 53BP1 (i, j). At least 1500 cells were acquired by wide-field microscopy, and γ-H2AX (i) or 53BP1 (j) foci were counted. The number of cells (n) for each condition is indicated. k, l Cells were immunostained with antibodies against PICH and UBTF, and the mean numbers of total and rDNA-associated UFBs per anaphase cell were monitored. For IF experiments, DNA was counterstained with DAPI. For UFB counting, at least 200 cells were acquired by wide-field microscopy. For all data bars, error bars represent ± SD from at least three independent experiments. The significance of differences was assessed in two-tailed unpaired Student’s t-tests. ***P < 0.0005, **P < 0.005, *P < 0.05, NS not significant

Fig. 5

CDA and MAPT levels are negatively correlated in cancer. Scatterplots showing the Pearson correlation between mRNA microarray data a–c or mRNA sequencing data d–g for CDA and MAPT expression for a Institut Curie breast cancer cell lines (n = 34), b CCLE (Cancer Cell Line Encyclopedia) cancer cell lines (n=967). c METABRIC (Molecular Taxonomy of Breast Cancer International Consortium) breast cancer tissues (n = 1866), and d–g TCGA (The Cancer Genome Atlas) four different cancer tissues. Dashed vertical lines correspond to mean CDA expression. P < 0.05 were considered statistically significant

Fig. 6

Schematic model of the interactions between CDA and Tau. a The simultaneous silencing of Tau and CDA increased genetic instability, as illustrated by the increases in γ-H2AX and 53BP1 focus events and UFB frequency. The accumulation of DNA damage and replication stress led to a decrease in clonal growth, and, thus, to cell death. b Tau downregulation affected rRNA synthesis by decreasing UBTF recruitment to rDNA promoters, thereby impairing the transcription of rDNA. By contrast, Tau overexpression in CDA-deficient cells led to an increase in rDNA transcription. Finally, Tau downregulation decreased the size of the intracellular ribonucleotide pools, consistent with previous reports linking decreases in rRNA production to imbalances in ribonucleotide pools