The human nucleoporin Tpr protects cells from RNA-mediated replication stress

Abstract

Although human nucleoporin Tpr is frequently deregulated in cancer, its roles are poorly understood. Here we show that Tpr depletion generates transcription-dependent replication stress, DNA breaks, and genomic instability. DNA fiber assays and electron microscopy visualization of replication intermediates show that Tpr deficient cells exhibit slow and asymmetric replication forks under replication stress. Tpr deficiency evokes enhanced levels of DNA-RNA hybrids. Additionally, complementary proteomic strategies identify a network of Tpr-interacting proteins mediating RNA processing, such as MATR3 and SUGP2, and functional experiments confirm that their depletion trigger cellular phenotypes shared with Tpr deficiency. Mechanistic studies reveal the interplay of Tpr with GANP, a component of the TREX-2 complex. The Tpr-GANP interaction is supported by their shared protein level alterations in a cohort of ovarian carcinomas. Our results reveal links between nucleoporins, DNA transcription and replication, and the existence of a network physically connecting replication forks with transcription, splicing, and mRNA export machinery.

Fig. 1. Tpr depletion sensitizes cells to replication stress.

a Clonogenic survival of Tpr-depleted (siTPR) and control U-2 OS cells (siCTRL) treated with aphidicolin (Aph). Data are represented as mean ± SEM from three independent experiments (n = 3). Statistical analysis: Brown–Forsythe and Welch ANOVA with Dunnett’s T3 post analysis for multiple comparisons; ns not significant; *p value < 0.05. b Clonogenic survival of Tpr-depleted (siTPR) and control U-2 OS cells (siCTRL) treated with hydroxyurea (HU). Data are represented as mean ± SEM from three independent experiments (n = 3). Statistical analysis: Mann–Whitney test and Welch’s t-test; ns not significant; **p value < 0.01. c Representative images of 53BP1 foci in Tpr-depleted (siTPR) and control (siCTRL) U-2 OS cells. Cells were either nontreated (Mock) or treated with 0.2 μM aphidicolin for 48 h (Aph). The staining of cyclin A serves as a marker of the cell cycle stage. Nuclei were stained with DAPI. The scale bar, 7.5 µm. Top right: quantification of 53BP1 foci in Tpr-depleted (siTPR) and control U-2 OS cells (siCTRL) either nontreated (Mock) or exposed to 0.2 μM aphidicolin for 48 h (Aph). A series of nonoverlapping frames were randomly acquired by automated microscopy, and the 53BP1 foci in G1 nuclei (cyclin A negative) were counted. Data are represented as mean ± SD from three independent experiments (n = 3). Statistical analysis: Kruskal–Wallis test with Dunnett’s T3 post analysis for multiple comparisons. Bottom right: quantification of micronuclei in Tpr-depleted (siTPR) and control U-2 OS cells (siCTRL) performed on the same samples as 53BP1 foci analysis (top right panel). Data are represented as mean ± SEM from three independent experiments (n = 3). Statistical analysis: Kruskal–Wallis test with Dunnett’s T3 post analysis for multiple comparisons. d Depletion of Tpr leads to increased RPA pS4/8 phosphorylation upon hydroxyurea treatment. Cells were transfected with three independent siRNAs targeting Tpr (siTPR #53, siTPR #54, siTPR #55) or control siRNA (siCTRL). Transfected cells were either nontreated (Mock) or treated with 2 mM hydroxyurea (HU) for 4 h. Staining of RPA total is used as a loading control. Data are representative of three independent experiments. Source data for d are provided in the Source Data file.

Fig. 2. Tpr depletion impacts replication fork progression.

a DNA fiber analysis of the replication fork speed and symmetry in nontreated conditions. Left: schematic of the CldU/IdU pulse-labeling protocol used. Middle: analysis of the replication fork speed. Total number of molecules analyzed: siCTRL (n = 541), siTPR (n = 567). The plot includes the mean ± SD. Statistical test: unpaired t-test with Welch’s correction, two-tailed. Right: analysis of the replication fork symmetry. The ratio between CldU/IdU was calculated from values in a (left). Whiskers indicate the 5th–95th percentiles (p value associated with Kolmogorov–Smirnov test). b DNA fiber analysis of the ability of forks to stop progression in the presence of replication stress. Top: schematic of the CldU/IdU pulse-labeling protocol used. Bottom: percentage of the double-labeled replication forks containing both CldU and IdU signal is showed. Total number of slides analyzed from three independent experiments: siCTRL (n = 6), siTPR (n = 9). The plot includes the mean ± SD. Statistical test: unpaired t-test with Welch’s correction, two-tailed. c DNA fiber analysis of the replication fork recovery upon HU removal. Left: schematic of the CldU/IdU pulse-labeling protocol used. Middle: length of the IdU tracks measured upon hydroxyurea removal. Total number of molecules analyzed: siCTRL (n = 297), siTPR (n = 310). The plot includes the mean ± SD. Statistical test: unpaired t-test with Welch’s correction, two-tailed. Right: analysis of the replication fork symmetry. The ratio between CldU/IdU was calculated from values in c (middle). Whiskers indicate the 5th–95th percentiles. Statistical test: unpaired t-test with Welch’s correction, two-tailed. d Categories of DNA replication intermediates identified in TEM analysis with a schematic representation of the molecule with dsDNA in black and ssDNA in red. Scale bars of 360 nm, which correspond to 1 kb of DNA, are reported on each picture. e Chart reporting the percentages of RIs identified. Gapped forks are defined as forks with ssDNA gaps at fork junction points longer than 200 nts. Data are represented as mean ± SD from four independent experiments (n = 4) for hydroxyurea-treated (HU) samples and two independent experiments (n = 2) for nontreated samples. Statistical analysis: two-way ordinary ANOVA followed by Sidak’s multiple comparison test.

Fig. 3. Tpr depletion leads to DNA–RNA hybrid accumulation.

a Tpr-depleted (siTPR) and control (siCTRL) cells. Treatments: nontreated (Mock); 2 mM hydroxyurea 4 h (HU); 50 μM cordycepin 3 h followed by 2 mM HU 4 h (HU + cord). RPA total, loading control. Data are representative of three independent experiments. Source data are provided in the Source Data file. b Quantification of γH2AX-foci in siTPR (red) and siCTRL (blue) HeLa cells. Data represent mean ± SEM from independent experiments (n = 3). Statistical analysis: two-tailed unpaired Student’ t test. c Quantification of 53BP1 foci in siTPR and siCTRL HeLa cells. Other details as in b. d DNA breaks in Tpr-depleted cells are transcription dependent. Alkaline comet assay in mock or cordycepin-treated (+Cord) siTPR (red) or siCTRL (blue) HeLa cells. Representative images (left) and mean ± SEM of median comet tail moment (right) from independent experiments (n = 3). Statistical analysis: two-tailed unpaired Student’ t test. e RNaseH1 expression rescues DNA breaks in Tpr-depleted HeLa cells. Alkaline comet assay in siTPR (red) and siCTRL (blue) cells expressing RNaseH1 from plasmid pEGFP-M27-H1 (+RNH1) or transfected with plasmid pEGFP-N2 (−RNH1). Other details as in d. f DNA–RNA hybrid immunoprecipitation with the S9.6 antibody in siTPR (red) and siCTRL (blue) HeLa cells at the APOE, RPL13A, EGR1, and BTBD19 genes. Where indicated samples were treated with RNase H (RNH) prior to immunoprecipitation. S9.6 signal values were normalized to the control cells value of each locus in each experiment. Location of the amplicon relative to each gene is indicated. Mean ± SEM of signal values of R-loop detection obtained from independent experiments are shown (n = 5, except for siTPR +RNH of RPL13A, siCTRL +RNH, and both siTPR of EGR1, siTPR −RNH of BTBD19 for which n = 4, and siTPR +RNH of BTBD19 for which n = 3). Statistical analyses with a two-tailed unpaired Student’ t test. g siTPR pool deconvolution analysis by S9.6 immunofluorescence in HeLa cells. Immunofluorescence with S9.6 and anti-nucleolin antibodies in siCTRL (blue), siTPR pool (siTPR pool; red), or each individual siRNAs of the siTPR pool (siTPR #1, siTPR #2, siTPR #3, siTPR #4; red) transfected cells. Representative images from three independent experiments (left) and quantification of S9.6 signal intensity per nucleus after the subtraction of the nucleolar signal (right). Scale bar, 7.5 µm. Median value and number of cells (n) examined over three independent experiments are indicated. Statistical analysis: two-tailed unpaired Mann–Whitney U test; ****p < 0.0001.

Fig. 4. Tpr maintains normal protein levels of GANP.

a U-2 OS cells were transfected using siRNAs and either nontreated (Mock) or treated with 2 mM hydroxyurea (HU) for 4 h. RPA, loading control. Data are representative of three independent experiments. b U-2 OS cells were transfected using siRNAs and treated as in a. RPA, loading control. The experiment was performed twice yielding similar results. c U-2 OS cells were transfected using siRNAs. Tubulin, loading control. Data are representative of three independent experiments. d U-2 OS cells were transfected using siRNA targeting endogenous Tpr (siTPR #54), GANP (siGANP #87), or control siRNA (siCTRL). In parallel, cells were cotransfected with siRNA-resistant Tpr-GFP insensitive to siTPR #54. Tpr-GFP and GANP protein levels were analyzed by immunoblotting. Tubulin, loading control. The experiment was performed twice yielding similar results. e U-2 OS cells were transfected with siRNAs and levels of Tpr and GANP were assessed using immunofluorescence. Nuclei were stained with DAPI. The scale bar represents 7.5 µm. Data are representative of three independent experiments. f Expression of siRNA-resistant full-length Tpr-wt can rescue GANP protein level. U-2 OS cells were transfected with siRNA targeting endogenous Tpr (siTPR), and with siRNA-resistant Tpr-wt-GFP (TPR-wt). Levels of Tpr-wt-GFP and GANP were assessed using immunofluorescence. Nuclei, DAPI. The scale bar, 7.5 µm. Data are representative of three independent experiments. g Expressions of Tpr-N-terminal domain or Tpr-C-terminal domain are not able to rescue GANP protein level. U-2 OS cells were transfected with siRNA targeting endogenous Tpr (siTPR), and with Tpr-N-terminal-FLAG (TPR-N) or Tpr-C-terminal-FLAG (TPR-C) constructs. Levels of Tpr-FLAG and GANP were assessed using immunofluorescence. Nuclei, DAPI. The scale bar, 7.5 µm. Data are representative of three independent experiments. h Expression of siRNA-resistant full-length Tpr-NEBM (Nuclear Envelope Binding Mutant) is not able to rescue GANP protein level. U-2 OS cells were transfected with siRNA targeting endogenous Tpr (siTPR), and with siRNA-resistant Tpr-NEBM1-GFP (NEBM1) construct carrying L458P/M489P double amino acid substitution in Tpr nuclear envelope binding domain. Levels of Tpr-NEBM1-GFP and GANP were assessed using immunofluorescence. Nuclei, DAPI. The scale bar, 7.5 µm. Data are representative of three independent experiments. Source data for a–d are provided in the Source Data file.

Fig. 5. GANP-depleted cells exhibit replication defects.

a DNA fiber analysis of the replication fork speed and symmetry in nontreated conditions. Left: schematic of the CldU/IdU pulse-labeling protocol used. Middle: analysis of the replication fork speed. Total number of molecules analyzed: siCTRL (n = 362), siGANP (n = 407). The plot includes the mean ± SD. Statistical test: unpaired t-test with Welch’s correction, two-tailed. Right: analysis of the replication fork symmetry. The ratio between CldU/IdU was calculated from the same molecules used for the analysis of the replication fork speed (middle) (siCTRL n = 362, siGANP n = 407). Bounds of box are 25th–75th percentile, center shows the median, whiskers indicate the 10th–90th percentiles, data outside the range are drawn as individual dots. Statistical analysis: unpaired t-test with Welch’s correction, two-tailed. b DNA fiber analysis of the replication fork recovery upon HU removal. Left: schematic of the CldU/IdU pulse-labeling protocol used. Middle: length of the IdU tracks measured upon hydroxyurea removal. Total number of molecules analyzed: siCTRL (n = 190), siGANP (n = 181). The plot includes the mean ± SD. Statistical test: unpaired t-test with Welch’s correction, two-tailed. Right: analysis of the replication fork symmetry. The ratio between CldU/IdU was calculated from the same molecules used for the analysis of the length of the IdU tracks measured upon hydroxyurea removal (middle) (siCTRL n = 190, siGANP n = 181). Bounds of box are 25th–75th percentile, center shows the median, whiskers indicate the 10th–90th percentiles, data outside the range are drawn as individual dots. Statistical analysis: unpaired t-test with Welch’s correction, two-tailed.

Fig. 6. Analysis of Tpr and GANP protein levels in human tumors.

a Immunohistochemical analysis of Tpr and GANP protein levels on parallel sections of human tumors. Archival specimens from a cohort of 51 human serous ovarian carcinomas were stained with antibodies specific for Tpr and GANP using immunoperoxidase technique. Examples of immunohistochemical staining patterns seen among the ovarian tumors are shown. The scale bar represents 100 µm. b The summary of the scoring categories for the entire cohort of 51 human serous ovarian carcinomas analyzed.