Lamin A/C depletion enhances DNA damage-induced stalled replication fork arrest

Abstract

The human LMNA gene encodes the essential nuclear envelope proteins lamin A and C (lamin A/C). Mutations in LMNA result in altered nuclear morphology, but how this impacts the mechanisms that maintain genomic stability is unclear. Here, we report that lamin A/C-deficient cells have a normal response to ionizing radiation but are sensitive to agents that cause interstrand cross-links (ICLs) or replication stress. In response to treatment with ICL agents (cisplatin, camptothecin, and mitomycin), lamin A/C-deficient cells displayed normal γ-H2AX focus formation but a higher frequency of cells with delayed γ-H2AX removal, decreased recruitment of the FANCD2 repair factor, and a higher frequency of chromosome aberrations. Similarly, following hydroxyurea-induced replication stress, lamin A/C-deficient cells had an increased frequency of cells with delayed disappearance of γ-H2AX foci and defective repair factor recruitment (Mre11, CtIP, Rad51, RPA, and FANCD2). Replicative stress also resulted in a higher frequency of chromosomal aberrations as well as defective replication restart. Taken together, the data can be interpreted to suggest that lamin A/C has a role in the restart of stalled replication forks, a prerequisite for initiation of DNA damage repair by the homologous recombination pathway, which is intact in lamin A/C-deficient cells. We propose that lamin A/C is required for maintaining genomic stability following replication fork stalling, induced by either ICL damage or replicative stress, in order to facilitate fork regression prior to DNA damage repair.

Fig 1

Functions of lamin A/C. (A) Ingenuity pathway analysis of lamin A/C based on mRNA microarray expression data from comparisons of mRNA from Lmna^+/+ to that of Lmna^−/− cells. Pink indicates upregulated and green downregulated genes. The top 12 relevant biological functions and disease associations from the analysis are shown on both sides of the figure. (B) Comparison of mRNA expression status between Lmna^+/+ and Lmna^−/− cells. The expression is organized into eight major groups involved in cellular metabolism. The eight major groups are cancer, proliferation and growth, cell cycle, cell death, signaling ligands and NF-κB, lipid metabolism, interferon signaling, and glutathione metabolism. (C) mRNA expression is organized from highest to lowest in Lmna^−/− cells compared to Lmna^+/+ cells. Details of the expression levels are available at http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE38777. (D) Western blot showing cyclin D1 protein levels in Lmna^+/+ and Lmna^−/− cells. (E) Western blot showing lamin A/C knockdown by LMNA-specific siRNA. (F) Western blot showing decreased cyclin D1 levels in 293 cells treated with lamin A/C-specific siRNA. (G) Coimmunoprecipitation of endogenous lamin A/C and cyclin D with anti-lamin A/C antibody detected by immunoblotting.

Fig 2

Ionizing radiation response in cells with and without lamin A/C. (A) Clonogenic survival of Lmna^+/+ and Lmna^−/− mouse cells after exposure to increasing radiation doses. (B) Western blot showing depletion of cyclin D1 by specific siRNA in Lmna^−/− and Lmna^+/+ cells. (C) Clonogenic survival of cells with and without cyclin D1 knockdown after exposure to graded doses of IR. (D) Exponentially growing cells were irradiated with 2 Gy, and the appearance/disappearance of γ-H2AX foci was determined by immunostaining. (E) Lmna^+/+ and Lmna^−/− cells were irradiated with 6 Gy, and 53BP1 focus formation was quantified by immunostaining at different time points postirradiation. (F) Exponential-phase Lmna^+/+ and Lmna^−/− cells were irradiated with different IR doses, and RAD51 focus formation was quantified 4 and 8 h postirradiation. (G) Ionizing radiation-induced phosphorylation of ATM Ser1981, detected by immunoblotting, in exponentially growing Lmna^+/+ and Lmna^−/− cells after irradiation with 5 Gy. (H) Ionizing radiation-induced phosphorylation of DNA-PK catalytic subunit (DNA-PKcs) in 293 cells with and without siRNA depletion of lamin A/C detected by DNA-PKcs and phospho-Ser2056 immunoblotting. Cells were irradiated with 10 Gy and collected at different time points postirradiation for analysis. (I) Chromosomal aberrations in Lmna^+/+ and Lmna^−/− cells after IR exposure. For analysis of G₁-phase aberrations, cells were irradiated (3 Gy), incubated for 12 h, and then treated for 3 h with colcemid before collecting metaphases for scoring. Categories of asymmetric chromosome aberrations scored included dicentrics, centric rings, interstitial deletions-acentric rings, and terminal deletions. For S-phase-specific aberrations, cells were irradiated with 2 Gy and incubated for 6 h, and metaphases were harvested after 3 h of colcemid treatment. For G₂-type chromosome aberrations, exponential-phase cells were irradiated (1 Gy) and incubated for 1 h, followed by 3 h of colcemid treatment to collect metaphases.

Fig 3

Clonogenic survival of Lmna^−/−, Lmna^+/+, and lamin A/C-depleted 293 cells treated with DNA-damaging agents. For survival assay, the required number of cells were plated, incubated for 6 h, treated with the indicated doses of drug or irradiated with UV, and incubated for 10 to 12 days to form countable colonies. Lmna^+/+ cells, Lmna^−/− cells, and Lmna^−/− cells with ectopic lamin A expression were treated with cisplatin (A), mitomycin C (B), camptothecin (C), hydroxyurea (D), formadehyde (F), VP16 (G), UV (H), and MNNG (I). (E) Surviving fraction of 293 cells with and without lamin A/C depletion with specific siRNA after HU treatment. *, P < 0.05; **, P < 0.01.

Fig 4

Impaired DNA damage response in lamin A/C-depleted cells. (A to C) Cells with γ-H2AX foci after treatment with cisplatin (A), MMC (B), and camtothecin (C). (D) Cells with FANCD2 foci after cisplatin and MMC treatment. (E) Frequency of Lmna^+/+ and Lmna^−/− metaphases with chromosome aberrations after cisplatin treatment. (F) Frequency of 293 metaphases with and without lamin A/C with chromosome aberrations after cisplatin treatment. (G) Frequency of Lmna^+/+ and Lmna^−/− metaphases with chromosome aberrations after MMC treatment. *, P < 0.05; **, P < 0.01.

Fig 5

Hydroxyurea treatment response in cells with and without lamin A/C. (A) Frequency of cells with γ-H2AX at different time points after treatment. (B) Lmna^+/+ and Lmna^−/− cells stained for DNA with DAPI and γ-H2AX immunostaining. EdU-positive S/G₂ cells (selected by DAPI staining) were selected only for γ-H2AX focus counting. (C) The number of larger γ-H2AX foci counted in late S/G₂-phase cells in Lmna^+/+ and Lmna^−/− cells represent collapsed forks. The quantification of EdU-positive late S/G₂-phase cells was performed by using scatter blots obtained by scanning the slides for DAPI and EdU intensity . γ-H2AX foci were counted in ∼40 cells per data point and experiment per slide. (D) Histogram showing percent EdU-positive G₁ cells. A total of 3,000 cells were counted, and the means from three experiments are plotted. *, P < 0.05; **, P < 0.01. (E to H) Frequency of cells with Mre11 (E), CtIP foci (F), RAD51 (G), RPA (H), and FANCD2 (I) foci after HU treatment. (J) HU-treated lamin A/C-depleted 293 cells at metaphase with chromosome aberrations, including a triradial (arrow). (K) Histogram showing a comparison of chromosome aberrations per metaphase in Lmna^−/− cells, Lmna^+/+ cells, and Lmna^−/− cells with ectopic expression of lamin A.

Fig 6

Effect of polymerase η depletion in cells with and without lamin A/C for DNA damage response. (A) Western blot showing polymerase η depletion with specific siRNA in Lmna^+/+ and Lmna^−/− cells. (B and C) Comparison of chromosome aberrations per metaphase in Lmna^+/+ and Lmna^−/− cells with and without polymerase η depletion after mitomycin (B) and HU (C) treatment. (D) Recruitment of polymerase η onto laser-induced DNA damage. Exponentially growing Lmna^+/+ and Lmna^−/− cells transfected with cDNA coding for YFP-polymerase η were microirradiated, and time-lapse images were captured. (E) YFP-polymerase η relative fluorescent intensity kinetics measured in Lmna^+/+ and Lmna^−/− cells after microirradiation. *, P < 0.05; **, P < 0.01; ***, P < 0.005.

Fig 7

Reinitiation of stalled DNA replication forks and initiation of new origins in Lmna^+/+ and Lmna^−/− cells. (A) Shown are DNA labeling and HU treatment protocol for single DNA fiber analysis (i) and three major types of labeled DNA tracts for analysis (ii). (B) Comparison of global DNA replication restart after release from 2 h of HU treatment. Cells were prelabeled with IdU, treated with HU, and then postlabeled with CldU, fixed, immunostained with IdU (green) and CldU (red) antibodies, and counterstained with DAPI (blue). Equal intensities of CldU and IdU as well as strong colocalization was observed in Lmna^+/+ MEFs, indicating that DNA replication was able to restart. In contrast, CldU staining in Lmna^−/− cells was very weak, and little colocalization was detected. (C) Quantification of percentages of cells with stalled forks after HU treatment. Lmna^−/− cells have the least incorporation of CldU and thus the maximum frequency of cells with stalled forks. (D) Representative images of replication tracks from Lmna^+/+ and Lmna^−/− cells after 1 h of HU treatment (i), after 21 h of HU treatment (ii), and in cyclin D1-depleted cells after 21 h of HU treatment (iii). (E) Quantification of stalled forks determined by fiber analysis with only IdU signal after 1 or 21 h of HU treatment. Lmna^−/− cells ectopically expressing lamin A were designated Lmna^−/− + lamin A. (F) Quantification of new origins as determined by CldU signal after 1 or 21 h of HU treatment. (G) Distribution of IdU track length from DNA fibers from control Lmna^+/+ and Lmna^−/− cells and Lmna^+/+ and Lmna^−/− cells treated with HU for 5 h.

Fig 8

Cyclin D1 is released from lamin A complexes after HU treatment. Cells were treated with 6 mM HU and collected at the indicated times. (A) Lamin A immunoprecipitates blotted for lamin A and cyclin D1. (B) Quantitation of cyclin D1 coimmunoprecipitated by lamin A antibody. The means and SD are from four independent experiments. (C) Effect of cyclin D1 depletion in Lmna^+/+ and Lmna^−/− cells on stalled forks after 1 or 21 h of HU treatment. (D) Effect of cyclin D1 depletion in Lmna^+/+ and Lmna^−/− cells on new origins by determining the CldU signal after 1 or 21 h of HU treatment. (E) Frequency of stalled forks in HU-treated Lmna^+/+ and Lmna^−/− cells with and without ectopic cyclin D1 expression, as determined by DNA fibers analysis of tracks labeled with IdU only. Percentages are based on the total number of IdU tracks that were counted in the different fields. (F) Quantification of new origins in HU-treated Lmna^+/+ and Lmna^−/− cells, with and without ectopic cyclin D1 expression.

Fig 9

Lamin A/C depletion does not affect DSB repair by homologous recombination. (A) Western blot analysis of lamin A/C depletion in MCF7 cells by siRNA. (B) HR frequencies in MCF7 cells are shown with or without I-SceI induction in untreated cells, in cells treated with control siRNA, and in cells treated with LMNA- or BRCA1-specific siRNA. The results presented are the means and standard errors from three independent experiments. (C) Telomere strand-specific orientation analysis at metaphase. Metaphase chromosome CO-FISH showing strand-specific telomeres was performed as described in Materials and Methods. (D) Lmna^+/+ and Lmna^−/− cells do not show any difference in telomeric circles. Genomic DNA from Lmna^+/+ and Lmna^−/− cells was separated by neutral-neutral 2D gel electrophoresis first in size (x axis) and then in shape (y axis), blotted, and probed for telomeric DNA. U2OS cells (ALT cells) were used as a positive control. (E) Model for the role of lamin A/C showing the start of homologous recombination (HR) repair at the proposed site for the collapse of stalled DNA replication forks in order to enable origins to restart and initiate homology-mediated repair.