Defective DSB repair correlates with abnormal nuclear morphology and is improved with FTI treatment in Hutchinson-Gilford progeria syndrome fibroblasts

E X P E R I M E N T A L C E L L R E S E A R C H 3 1 6 ( 2 0 1 0 ) 2 7 4 7 – 2 7 5 9

ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r . com/ loca te /yexc r

Research Article

Defective DSB repair correlates with abnormal nuclearmorphology and is improved with FTI treatment inHutchinson-Gilford progeria syndrome fibroblasts

Dan Constantinescub,c, Antonei B. Csokad, Christopher S. Navaraa,c, Gerald P. Schattena,b,c,⁎aDivision of Developmental and Regenerative Medicine, Department of Obstetrics, Gynecology, and Reproductive Sciences,University of Pittsburgh, Pittsburgh, PA 15260, USAbDepartment of Cell Biology-Physiology, University of Pittsburgh, Pittsburgh, PA 15260, USAcPittsburgh Development Center, Magee-Women's Research Institute, University of Pittsburgh, Pittsburgh, PA 15260, USAdDivision of Geriatrics, Department of Medicine, University of Pittsburgh Medical Center, Pittsburgh, PA 15260, USA

A R T I C L E I N F O R M A T I O N

⁎ Corresponding author. 204 Craft Ave., B609, PE-mail address: [email protected] (G.P

0014-4827/$ – see front matter © 2010 Publisdoi:10.1016/j.yexcr.2010.05.015

A B S T R A C T

Article Chronology:

Received 30 September 2008Revised version received 6 May 2010Accepted 18 May 2010Available online 25 June 2010

Impaired DSB repair has been implicated as a molecular mechanism contributing to theaccelerating aging phenotype in Hutchinson-Gilford progeria syndrome (HGPS), but neither theextent nor the cause of the repair deficiency has been fully elucidated. Here we perform aquantitative analysis of the steady-state number of DSBs and the repair kinetics of ionizingradiation (IR)-induced DSBs in HGPS cells. We report an elevated steady-state number of DSBs andimpaired repair of IR-induced DSBs, both of which correlated strongly with abnormal nuclearmorphology. We recreated the HGPS cellular phenotype in human coronary artery endothelialcells for the first time by lentiviral transduction of GFP-progerin, which also resulted in impairedrepair of IR-induced DSBs, and which correlated with abnormal nuclear morphology. Farnesyltransferase inhibitor (FTI) treatment improved the repair of IR-induced DSBs, but only in HGPS

cells whose nuclear morphology was also normalized. Interestingly, FTI treatment did not result ina statistically significant reduction in the higher steady-state number of DSBs. We also report adelay in localization of phospho-NBS1 and MRE11, MRN complex repair factors necessary forhomologous recombination (HR) repair, to DSBs in HGPS cells. Our results demonstrate acorrelation between nuclear structural abnormalities and the DSB repair defect, suggesting amechanistic link that may involve delayed repair factor localization to DNA damage. Further, ourresults show that similar to other HGPS phenotypes, FTI treatment has a beneficial effect on DSBrepair.

© 2010 Published by Elsevier Inc.

Keywords:

AgingDNA repairProgeriaNuclear laminaLamin A

Introduction

Hutchinson-Gilford progeria syndrome (HGPS) is a rare geneticdisorder that causes rapid premature aging shortly after birth,recapitulating multiple pathologies associated with aging and

ittsburgh, PA 15232, USA.. Schatten).

hed by Elsevier Inc.

resulting in a median lifespan of 13 years (reviewed in Pollex andHegele [1]). The disease is caused by de novo mutations withinexon 11 of the LMNA gene which partially activates a cryptic splicedonor site, resulting in deletion of 50 amino acids from exon 11,with subsequent production of a “truncated” form of lamin A

Fax: +1 412 641 2410.

http://dx.doi.org/10.1016/j.yexcr.2010.05.015

mailto:[email protected]


2748 E X P E R I M E N T A L C E L L R E S E A R C H 3 1 6 ( 2 0 1 0 ) 2 7 4 7 – 2 7 5 9

termed, progerin or lamin AΔ50 [2,3]. LMNA encodes the A-typenuclear lamins, primarily lamins A and C. Along with the primaryB-type lamins, lamin B1 and lamin B2, the A-type lamins form thenuclear lamina [4,5], a scaffold-like structure that lines the innernuclear membrane and also contributes to the nuclear matrix [6].Multiple LMNAmutations produce nuclear structural irregularities,demonstrating that A-type lamins are intricately involved innuclear structural organization (reviewed in Dechat et al. [7]).Furthermore, there are multiple lines of evidence indicating acentral role for A-type lamins in chromatin organization (reviewedin Dechat et al. [7]).

Accordingly, HGPS cells exhibit altered nuclear structuralcharacteristics and chromatin organization. Nuclear structuralirregularities include changes in the spatial distribution of nuclearpore complexes [8] as well as modifications of the nuclear lamina[8,9] that are likely responsible for the abnormal nuclearmorphology observed in primary HGPS cells [2,3,8]. Thesestructural irregularities are accompanied by functional changes,including reduced deformability of the nuclear lamina [9],increased nuclear stiffness and sensitivity to mechanical stress[10], and mitotic defects, including abnormal chromosomesegregation, delays in cytokinesis, nuclear reassembly, andbinucleation [11,12]. Multiple chromatin organization changeshave been described as well. Particularly, HGPS cells exhibit loss ofperipheral heterochromatin [8,13] that may be due to epigeneticchanges, including up-regulation of H3K9me3 and H4K20me3,which define constitutive heterochromatin, and downregulationof H3K27me3, which defines facultative heterochromatin [13–15].In addition to epigenetic changes, HP1α, which is usuallyassociated with H3K9me3, is downregulated and partially disso-ciated [14,15]. It is likely that these abnormalities influencenuclear biological processes that are dependent on proper nuclearstructure and chromatin organization.

It has been demonstrated that the nuclear morphologicalchanges and altered chromatin characteristics are not caused bylamin A haploinsufficiency but by the presence of progerin in adominant gain-of-function fashion, possibly through its accumu-lation at the inner nuclear membrane (INM) [8,14], where itappears to alter nuclear lamina structure [9]. Accordingly,reduction of the farnesylated form of progerin using farnesyltransferase inhibitors (FTIs) [16–22] and reduction of progerinusing antisense morpholinos [14] improve nuclear abnormalitiesin HGPS cells as well as various disease phenotypes in HGPSmousemodels. The success of FTIs in these studies and the current lack ofany other therapeutic approach for HGPS have lead to a currentphase II clinical trial examining the beneficial effect of FTIs in HGPSpatients (ClinicalTrials.gov identifier: NCT00425607).

Interestingly, recent studies suggest that DSB accumulation dueto impaired DSB repair is one of the mechanisms leading to theaccelerated aging phenotype [19,23–25]. DSB accumulationappears to be due at least in part to impaired localization of DSBrepair factors including Rad51 and Rad50 [24]. It has also beenshown that xeroderma pigmentosum group A (XPA), a nucleotideexcision repair protein (NER), aberrantly localizes to a subset ofDSBs in HGPS cells through interaction with chromatin andinhibits the localization of Rad51 and Rad50, perhaps throughsteric hindrance [24]. There is evidence to suggest that XPA doesnot localize to camptothecin (CPT)-induced DSBs, indicating thatXPA-localized DSBs may be functionally different in origin orrepair [24]. However, XPA-mediated interference with DSB repair

does not fully explain the DSB repair problem in HGPS cells sincerepair of CPT-induced DSBs in HGPS cells is slower compared towild type cells [24].

In this study, we performed a quantitative analysis of thesteady-state number of DSBs and the repair kinetics of IR-induced DSBs in HGPS fibroblasts. We also examined whetherthe observed deviations from wild type for these characteristicscorrelated with abnormal nuclear morphology and whether theywere caused by progerin. Further, we examined whether FTItreatment can decrease the steady-state DSB level and improvethe repair kinetics of IR-induced DSBs in HGPS cells. Lastly, toexpand on previous studies that showed that localization ofrepair factors to DSBs is impaired, we performed a quantitativeanalysis of the localization kinetics of the MRN repair complexfactors, phospho-NBS1 and MRE1, to γ-irradiation-inducedDSBs.

Materials and methods

Cell culture

Primary dermal fibroblasts from HGPS donors (AG11513 andAG11498) and from apparently healthy donors (AG08470 andAG16409) (Coriell Cell Repository, Camden, NJ) were grown inDulbecco's modified Eagle's medium (DMEM) supplemented with15% fetal bovine serum, 1mM L-glutamine, 1% Pen/Strep, 1% MEMnonessential amino acids. Cells were passaged every 4–6 days, andmedium was changed every 2 days. Primary human coronaryartery endothelial cells (Cambrex, East Rutherford, NJ) werecultured in endothelial cell basal medium (EBM) supplementedwith the EGM-MV bullet kit (Cambrex). Cells were passaged every6–8 days, and medium was changed every 2 days.

FTI treatment

Cells were passaged, and the medium was replaced 24 hours laterwith media containing 5 μM FTI L-744832 (Biomol, PlymouthMeeting, PA). FTI media was replaced every 24 hours for 72 hoursas described. For irradiated cells, treatment was continued every24 hours after irradiation until cells were assayed.

Western analysis

Western immunoblotting was performed according to establishedprotocols. Briefly, 10 μg of protein was loaded onto Tris–HCl 4–15%linear gradient gels (Bio-Rad Laboratories, Hercules, CA), andindividual gelswere run at 60 mA, 200 V. Proteinswere transferredto PDVF membrane for 1 hour at 350 mA, 100 V. The PVDFmembrane was blocked for 1 hour in Tris-buffered saline (TBS)containing 5% nonfat dry milk and 0.1% Tween-20. The followingprimary antibodies were diluted in blocking solution and used asneeded: mouse anti-human MRE-11 monoclonal antibody used at1:500 (Novus Biologicals, Littleton, CO) and rabbit anti-humanNBS1 used at 1:5000 (Millipore, Billerica,MA). Anti-mouse (ZymedLaboratories, San Francisco, CA) and anti-rabbit (InVitrogen,Carlsbad, CA) HRP-conjugated goat secondary antibodies wereused at 1:50,000 and 1:15,000, respectively, and detected viachemiluminescence using an ECL Advance Western BlottingDetection Kit (GE Healthcare, Buckinghamshire, UK).

2749E X P E R I M E N T A L C E L L R E S E A R C H 3 1 6 ( 2 0 1 0 ) 2 7 4 7 – 2 7 5 9

Immunocytochemistry

Briefly, cells were passaged onto glass slides and fixed using−20 °C methanol or 2% paraformaldehyde, as appropriate. Thecells were washed 2× in PBS and 1× in 0.1 % Triton-X PBS andblocked for 30 minutes in BSA (0.3%)/goat serum (5%) at roomtemperature. Cells were then incubated with the followingprimary antibodies as needed for one hour at room temperature:rabbit anti-human (pS353) polyclonal antibody, mouse anti-humanMRE-11monoclonal antibody, rabbit anti-human γ-H2AXpolyclonal antibody (Novus Biologicals, Littleton, CO), and mouseanti-human γ-H2AX monoclonal antibody (Upstate, Lake Placid,NY). The cells were washed 2× in PBS and 1× in 0.1% Triton-X PBS.The cells were incubated with anti-mouse and anti-rabbit AlexaFluor-488, -568, and -633 secondary antibodies as appropriate(Molecular Probes, Eugene, OR). DNAwas visualized with TOTO-3as needed (Molecular Probes Inc, Eugene, OR). The cells werewashed 2× in PBS and 1× in 0.1 % Triton-X PBS andmounted usingVectaShield mountingmedium (Vector Laboratories, Burlingame,CA). Confocal microscopy was performed with a Leica TCS SP2confocal microscope.

Construction of GFP-lamin and GFP-progerinlentiviral vectors

A human full-length LMNA Expressed Sequence Tag (EST)(GenBank accession number BC014507) was purchased (InVitro-gen, Carlsbad, CA). Using the EST as template, the coding region ofthe LMNA gene was amplified by PCR using the forward primer 5′-GCCACCATGGAGACCCCGTCCCAGCG-3′ and reverse primer 5′-TTACATGATGCTGCAGTTCTGGGGGCT-3′, and then subcloned intothe vector PCR2.1 (InVitrogen). A progerin coding sequence wasalso created directly from the same EST using a novel site-directedmutagenesis method in which the 150 nucleotide deletion ofthe progerin mRNA was introduced by PCR using the same for-ward primer as before, but with a reverse primer of sequence5 ′ -TTACATGATGCTGCAGTTCTGGGGGCTCTGGGCTCCT-GAGCCGCTGGCAGATGCCTTGTCGGCAG-3′. The LMNA and pro-gerin open reading frames (ORFs) were then subcloned from thePCR2.1 vector into the pEGFP-C1 vector (Clontech, Palo Alto, CA)using EcoRI digestion followed by ligation. This fused the LMNA/progerin cDNAs to the C-terminus of green fluorescent protein (GFP).The GFP-LMNA and GFP-progerin fusion protein ORFs were finallysubcloned into the lentiviral vector c-FUWbydigestionwithNheI andXbaI and subsequent ligation to create the vectors cFUW-GFP-LMNAand cFUW-GFP-progerin. The c-FUW vector drives constitutiveexpression in mammalian cells from the ubiquitin-C promoter [26].

Lentivirus production and transduction

GFP-LMNA and GFP-progerin lentivirus was produced using theViraPowerTM Lentivirus Expression System following the manu-facturer's instructions (InVitrogen, Carlsbad, CA). Briefly, 293FTcells were simultaneously transfected with ViraPower PackagingMix and either c-FUW-GFP-LMNA or c-FUW-GFP-progerin usingSuperfect transfection reagent (InVitrogen) according to themanufacturer's instructions (The ViraPower Packaging Mix con-tains the necessary structural and replication components in transon three separate plasmids). The c-FUW-GFP-LMNA and c-FUW-GFP-progerin lentiviral plasmids contain elements that allow

packaging of the gene constructs into virions. Supernatant wascollected 60 hours after transfection, and the virus was concen-trated via ultracentrifugation at 50,000×g for 3 hours. Cells weretransduced at an MOI of 20 for 18 hours. Transfected cell cultureswere passaged twice. Heterogeneous cell cultures were used forexperiments after the second passage.

Results

HGPS and wild-type fibroblasts with an abnormal nuclearmorphology have elevated steady-state levels of DSBs

To investigate DNA damage repair in HGPS cells, we examinedprimary dermal fibroblasts derived from two HGPS patients(AG11513 and AG11498) and two apparently healthy individuals(AG08470 and AG16409) of similar age (Supplemental Fig. 3A).Since there is evidence that HGPS cellular phenotypes increase inseverity with increasing passage number [8], we used HGPS andwild-type cells of similar population doubling (SupplementalFig. 3A). The HGPS fibroblasts used in these experiments displayclassical HGPS phenotypes including decreased proliferative capac-ity (Supplemental Fig. 1A), abnormal nuclear morphology (Supple-mental Fig. 1B), and increased (SA) β-galactosidase activity,indicating a higher percentage of senescent cells (SupplementalFigs. 1C and D).

To examine the difference in the steady-state number of DSBsbetween HGPS and wild-type fibroblasts, we quantified the averagenumber of DSBs per cell using the γ-H2AX foci assay. H2AX isphosphorylated specifically atDSBswithinminutes of formation andis dephosphorylated after repair [27–30]. The formation anddisappearance of γ-H2AX foci has been routinely used as a measureof DSB formation and repair, respectively [23–25,27–30]. A popula-tion of HGPS fibroblasts consists of cells that exhibit the character-istic abnormal nuclear morphology associated with HGPS,characterized by lobes and invaginations in the nuclear lamina, aswell as cells that exhibit a normal nuclear morphology [8]. Toexaminewhether potential abnormalities in the steady-state level ofDSBs correlate with the abnormal nuclear morphology, we visuallyseparated HGPS fibroblasts into two categories based on nuclearstructure using immunocytochemistry for lamin A: normal (HGPSN)and abnormal (HGPSAB) cells. Further, about 10% of AG08470 andAG16409 wild-type fibroblasts display an abnormal nuclear mor-phology that appears similar to that of HGPS cells (Supplemental Fig.1B). Although the abnormal nuclearmorphology inwild-type cells isdisease-independent, we distinguished between wild-type cellswith a normal nuclear morphology (wild-typeN) and those with anabnormal nuclear morphology (wild-typeAB) to examine whethergeneral nuclear structural abnormalities correlate with changes inthe steady-state level of DSBs. Representative images of wild-typeand HGPS cells exhibiting normal and abnormal nuclear structureare visualized in Fig. 1Ausing immunofluorescent labelingof laminAand γ-H2AX. Averaged together, total HGPS fibroblasts (HGPST)exhibited a 2.7-fold great average number of γ-H2AX foci per cellcompared to total wild-type fibroblasts (wild-typeT) (Fig. 1B andSupplemental Fig. 3B), indicating an elevated level of steady-stateDSBs. Examination of cells according to nuclear morphologyrevealed that both wild-type (AG08470AB and AG16409AB) andHGPS (AG11513AB and AG11498AB) fibroblasts with an abnormalnuclear morphology had significantly higher average numbers of


Fig. 1 – HGPS and wild-type fibroblasts with an abnormal nuclear morphology have elevated steady-state levels of DSBs.Immunocytochemistry was performed for γ-H2AX and lamin A in wild-type and HGPS fibroblasts. (A) Representative confocalimmunofluorescent images of γ-H2AX foci (red), lamin A (green), and DNA (blue), showing wild-type and HGPS fibroblast withnormal or abnormal nuclear morphologies. AG08470N and AG16409N are wild-type fibroblasts with a normal nuclear morphology,and AG08470AB and AG16409AB are wild-type cells with an abnormal nuclear morphology. AG11513N and AG11498N are HGPSfibroblasts with a normal nuclear morphology. AG11513AB and AG11498AB are HGPS fibroblasts with an abnormal nuclearmorphology. (B) γ-H2AX foci were visually detected and the average number of foci per cell was calculated. Total=all cellsincluding those with normal and abnormal nuclear morphologies. Normal=cells with a normal morphology. Aabnormal=cellswith an abnormal morphology. Data are from three different experiments. At least 100 cells were examined per time point. Forstatistical analysis, cells from multiple experiments were pooled together for each experimental group, and a one-way ANOVA testwas performed followed by a Tukey post-test to obtain p values for the differences between specific pairs of means. Error barsrepresent the standard deviation (SD) of the number of γ-H2AX foci per cell to highlight the large variance from cell to cell. Confocalimages were captured using a Leica TCS SP2 using Leica software. Scale bar equals 10 μm. *p<0.05, **p<0.01, and ***p<0.001.

γ-H2AX foci per cell compared to wild-type (AG08470N andAG16409N) and HGPS (AG11513N and AG11498N) fibroblastswith normal nuclear morphology (Figs. 1B and SupplementalFig. 3B). In particular, wild-typeAB fibroblasts had a 3.2-foldgreater average number of γ-H2AX foci per cell compared towild-typeN fibroblasts. Similarly, HGPSAB fibroblasts had a 5.2-foldgreater average number of γ-H2AX foci per cell compared toHGPSN fibroblasts. Interestingly, HGPSAB fibroblasts had a higheraverage number of γ-H2AX foci per cell compared to wild-typeABfibroblasts. No statistically significant difference was observedbetween wild-typeN and HGPSN fibroblasts, indicating that

HGPSN cells have a normal steady-state level of DSBs. Therewas also no statistically significant difference between wild-typeN and wild-typeT cells, most likely due to the fact that 90% ofthe total wild-type cell population is wild-typeN. Because of thesimilarity in the average number of DSBs between wild-typeNand wild-typeT, we did not distinguish between wild-typeN andwild-typeAB for subsequent experimentation and instead com-pared HGPSN and HGPSAB cells to wild-typeT cells. Altogether,these results indicate that HGPS fibroblasts have an elevatedaverage level of steady-state DSBs compared to wild-type cellsand demonstrate a correlation in both cell types between


abnormal nuclear morphology and elevated steady-state DSBlevels.

HGPS fibroblasts with an abnormal morphology displayimpaired repair of IR-induced DSBs

The higher steady-state level of DSBs in HGPSAB fibroblasts could bedue to an increased rate in DSB formation or to impaired DSB repair.In order to test the latter hypothesis, we performed the γ-H2AX fociassay at 0.5, 2, 6, 12, 24, and 48 hours after irradiation to determinethe kinetics of repair (Fig. 2). As for unirradiated cells, we visuallydistinguished between HGPSN (Fig. 2A, middle two columns) andHGPSAB fibroblasts (Fig. 2A, right two columns), but examinedwild-typeT fibroblasts (Fig. 2A, left two columns). Representative imagesat particular time points are visualized in Fig. 2A using immunocy-tochemistry for lamin A and γ-H2AX (lamin A labeling is omittedfrom the images in order to better visualize the γ-H2AX foci). Theaverage number of γ-H2AX foci per cell was quantified in Fig. 2B. At0.5 hours after irradiation (Fig. 2B and Supplemental Fig. 3C), whichhas been previously reported to reflect the apex of γ-H2AX fociformation and represent the maximal number of DSBs formed[27,28,31], HGPSAB fibroblasts exhibited no statistically significantdifference compared to wild-type and HGPSN fibroblasts, indicatingthat H2AX phosphorylation occurs properly (Figs. 2B and Supple-mental Fig. 3C). However, at 2, 6, 12, 24, and 48 hours afterirradiation, HGPSAB fibroblasts had a 1.5-fold, 3.1-fold, 2.7-fold, 3.9-fold, and 6.5-fold higher average number of γ-H2AX foci per cellcompared to wild-type fibroblasts, respectively (Figs. 2B andSupplemental Fig. 3C). There was no statistically significantdifference in the average number of γ-H2AX foci between HGPSNandwild-type fibroblasts at any time point after irradiation (Figs. 2Band Supplemental Fig. 3C). Further, whereas wild-type and HGPSNfibroblasts exhibited a return to steady-state DSB levels within24 hours, HGPSAB fibroblasts still exhibited statistically significantdifferences even at 48 hours after irradiation (Supplemental Fig. 2).These results indicate that repair of IR-inducedDSBs is compromisedspecifically in HGPS cells with an abnormal nuclear morphology,highlighting a correlation between the two phenotypes.

Elevated steady-state levels of DSBs and impaired repair ofIR-induced DSBs are detected in wild-type cells afterGFP-progerin expression

The abnormal nuclear morphology observed in HGPS cells iscaused by the presence of progerin in a dominant gain-of-functionfashion [8,15]. To examine whether progerin is fundamentallyresponsible for the elevated steady-state number of DSBs and theimpaired repair of IR-induced DSBs in HGPS cells, we expressedGFP-progerin in wild-type hCAECs, a cell type intimately involvedin the accelerated atherosclerosis so prevalent in HGPS, usinglentiviral transduction (Fig. 3A, right columns). We also expressedGFP-LMNA to control for overexpression of a lamin A-like protein(Fig. 3A, middle left column). As far as we know, this is the firsttime the HGPS phenotype has been recreated in human coronaryartery cells. Expression of GFP-progerin resulted in GFP-progerin-positive hCAECs with normal (Fig. 3A, middle right column) orabnormal nuclear morphologies (Fig 3A, right column) similar towhat is observed in HGPS fibroblasts. Expression of GFP-LMNA hadlittle to no effect on nuclear morphology (Fig. 3A, middle leftcolumn), similar to previous reports [8]. As before, we categorized

GFP-progerin-positive hCAECs as having normal (GFP-HGPSN)or abnormal (GFP-HGPSAB) nuclear morphology. UnirradiatedGFP-HGPSAB cells exhibited a 4.9-fold higher average number ofγ-H2AX foci per cell compared to wild-type cells (Fig. 3B andSupplemental Fig. 3D), indicating that expression of GFP-progerinincreases the higher steady-state number of DSBs. Analysis of therepair of IR-induced DSBs shows that expression of GFP-progerinin wild-type hCAECs compromises DSB repair. At 0.5 hours afterirradiation, all cell types had a similar average number of γ-H2AXfoci per cell (Fig. 3B and Supplemental Fig. 3D), indicating thatGFP-progerin has no effect on H2AX phosphorylation. At 12 and24 hours after irradiation GFP-HGPSAB cells exhibited a 2-fold and4.3-fold higher average number of γ-H2AX foci per cell comparedto wild-type cells (Fig. 3B and Supplemental Fig. 3D), indicatingthat GFP-progerin interferes with repair of IR-induced DSBs. Therewas no statistically significant difference between wild-type andGFP-HGPSN cells (Fig. 3B and Supplemental Fig. 3D), indicatingthat steady-state DSB levels were elevated and DSB repair wascompromised only in cells whose nuclear morphologywas altered.Further, no statistically significant differences were detectedbetween wild-type and GFP-LMNA-positive cells (Fig. 3B andSupplemental Fig. 3D), indicating that overexpression of lamin Ahad little or no effect onDSB repair. Altogether, these results indicatethat progerin is fundamentally responsible for the elevated steady-state levels of DSBs and defective repair of IR-induced DSBs in HGPSin a dominant gain-of-function manner and that both phenotypesstrongly correlate with the abnormal nuclear morphology.

FTI treatment improves nuclear morphology and repair ofIR-induced DSBs but does not reduce the steady-state level ofDSBs in HGPS fibroblasts

FTI treatment, which inhibits thematuration of progerin, improvessome of the nuclear abnormalities associated with HGPS both invitro and in animal models [16–22]. To examine whether FTItreatment reduces steady-state DSB levels, we quantified theaverage number of γ-H2AX foci per cell after 72 hours oftreatment. To examine whether FTI treatment improves the repairkinetics of IR-induced DSBs, we quantified the average number ofγ-H2AX foci per cell at 0.5, 24, and 48 hours after irradiation. Forthis purpose, FTI treatment was initiated 72 hours before irradi-ation and continued until the cells were assayed with cellsexamined 24 and 48 hours after irradiation receiving a total of96 and 120 hours of FTI treatment, respectively.

First, FTI treatment resulted in a statistically significant reductionin the percentage of HGPS fibroblasts with an abnormal nuclearmorphology, as determined using immunofluorescence microscopyto visualize lamin A/C (Fig. 4A and Supplemental Fig. 3E), similar toprevious reports [16,17,22]. The decrease in the percentage of cellswith an abnormal morphology correlated with the length of FTItreatment. Unirradiated FTI-treated AG11513 and AG11498 HGPSfibroblasts (72 hours total treatment) exhibited a 36% and 41%reduction in the percentage of cells with an abnormal nuclearmorphology, respectively, compared to nontreated cells (Fig. 4A,white bars). FTI-treated AG11513 and AG11498 fibroblasts exam-ined 24 hours after irradiation (96 hours total treatment) showed a39% and 44% reduction, respectively (Fig. 4A, gray bars). FTI-treatedAG11513 and AG11498 fibroblasts examined 48 hours after irradi-ation (120 hours total treatment) showed a 49% and 47% reduction,respectively (Fig. 4A, black bars). The percentage of FTI-treatedHGPS


Fig. 2 – HGPS fibroblasts with an abnormal morphology display impaired repair of IR-induced DSBs. Immunocytochemistry wasperformed for γ-H2AX and lamin A in 0.5, 2, 6, 12, 24, and 48 hours after 2 Gy ionizing radiation. (A) Representative confocalimmunofluorescent images ofγ-H2AX foci (red) and DNA (blue) at representative timepoints after irradiation show easily identifiableγ-H2AX foci. Lamin A labeling was omitted for presentation to better visualize the γ-H2AX foci. AG08470T and AG16409T are totalwild-type fibroblasts. AG11513N and AG11498N are HGPS fibroblasts exhibiting normal nuclear morphology. AG11513AB andAG11498AB are HGPS fibroblasts exhibiting abnormal nuclear morphology. (B) γ-H2AX foci were visually detected, and the averagenumber of foci per cell was calculated for the indicated timepoints. Data are from three different experiments. At least 100 cells wereexamined per time point. For statistical analysis, cells frommultiple experiments were pooled together for each experimental group,and a one-way ANOVA test was performed followed by a Tukey post-test to obtain p values for the differences between specific pairs ofmeans. Errorbars represent the standarddeviation(SD)of thenumberofγ-H2AX foci per cell tohighlight the large variance fromcell tocell. Confocal images were captured using a Leica TCS SP2 using Leica software. Scale bar equals 10 μm. *p<0.05, **p<0.01, and***p<0.001.

image of Fig.�2


Fig. 3 – GFP-progerin expression increases the steady-state level of DSBs and impairs repair of IR-induced DSBs in normal hCAECs.(A) Immunocytochemistry was used to examine wild-type, GFP-LMNA-positive, GFP-HGPSN-positive (normal nuclear morphology),and GFP-HGPSAB-positive (abnormal nuclearmorphology) hCAECs for γ-H2AX (red). GFP fluorescence was captured (green). Stableexpression of the GFP constructs was achieved using lentiviral transduction. Only GFP-progerin-positive cells were examined.(B) γ-H2AX foci were visually detected, and the average number of foci per cell was calculated for the indicated time points. Data arefrom two different experiments. At least 50 cells were examined per time point. For statistical analysis, cells from multipleexperiments were pooled together for each experimental group, and a one-way ANOVA test was performed followed by a Tukeypost-test to obtain p values for the differences between specific pairs of means. Error bars represent the standard deviation (SD) ofthe number of γ-H2AX foci per cell to highlight the large variance from cell to cell. Confocal images were captured using a Leica TCSSP2 using Leica software. Scale bar equals 10 μm. *p<0.05, **p<0.01, and ***p<0.001.

image of Fig.�3


fibroblasts with abnormal nuclear morphology remained statistical-ly higher than wild-type fibroblasts for unirradiated cells as well ascells examined after irradiation, indicating that FTI treatment doesnot restore normal morphology to all HGPS fibroblasts in the timeperiods examined, consistent with other reports [16,17,22].

To analyze the effect of FTI treatment on the steady-state levelof DSBs and repair of IR-induced DSBs in HGPS cells, we used totalHGPS (HGPST) cells instead of distinguishing between those withnormal and abnormal morphology because FTIs improve nuclearmorphology, which we have shown to correlate with the DSB

image of Fig.�4


repair defect and thus, may cause erroneous interpretation of theresults. First, grouping HGPS cells with normal and abnormalmorphologies together still resulted in a statistically significanthigher average number of γ-H2AX foci per cell compared to wild-type cells, for both unirradiated cells and cells examined 24 and48 hours after irradiation (Fig. 4B and Supplemental Fig. 3F).Averaged together, HGPST fibroblast cultures exhibited a 3.2-foldhigher steady-state average number of γ-H2AX foci per cellcompared to wild-type cultures (Fig. 4B and Supplemental Fig. 3F).After irradiation, no statistically significant differences wereobserved at 0.5 hours between HGPST and wild-type cultures,but HGPST cultures did exhibit 2.15-fold and 2.3-fold higheraverage numbers of γ-H2AX foci per cell at 24 and 48 hours afterirradiation, respectively (Fig. 4B and Supplemental Fig. 3F).

Quantification of the average number of γ-H2AX foci per cell forunirradiated FTI-treated HGPST (AG11513TFTI and AG11498TFTI)fibroblasts resulted in a statistically insignificant reduction com-pared to untreated HGPST (AG11513T and AG11498T) fibroblasts(Fig. 4B and Supplemental Fig. 3F). At 0.5 hours after irradiation,there was no statistically significant difference in the averagenumber of IR-induced γ-H2AX foci between FTI-treated andnontreated HGPST fibroblasts (AG11513T and AG11498T) (Fig. 4Band Supplemental Fig. 3F). However, at 24 and 48 hours afterirradiation, FTI-treated HGPST fibroblasts (AG11513TFTI andAG11498TFTI) exhibited a statistically significant reduction in theaverage number of IR-induced γ-H2AX foci per cell compared tonontreated HGPST fibroblasts (AG11513T and AG11498T) (Fig. 4Band Supplemental Fig. 3F). The differences in the average number ofγ-H2AX foci per cell for FTI-treated HGPS fibroblasts (AG11513TFTIand AG11498TFTI) and wild-type fibroblasts were not statisticallysignificant at 24 and 48 hours after irradiation under our experi-mental conditions (Fig. 4B and Supplemental Fig. 3F), indicating thatFTI treatment improves repair of IR-induced DSBs in HGPSTfibroblasts to an efficiency that is indistinguishable from wild-typecells. Altogether, these results indicate that although FTI treatmentcan improve repair of IR-induced DSBs, it may not be able to reducethe steady-state level of DSBs in HGPS cells.

We hypothesized that the improvement in repair of IR-inducedDSBs in HGPST cells may represent an improvement in all cells

Fig. 4 – FTI treatment improves nuclear morphology and repair ofsteady-state level of DSBs. HGPS cells and cells from healthy individuin irradiation studies were treated with the FTI after irradiation unand lamin A at the indicated time points after 2 Gy of γ-irradiationpostirradiation (gray bars), and 48 hours postirradiation (black barspresence (+) or absence (−) of FTI. Data are from experiments. Atrepresent the standard error of themean (SEM). (B) Examination ofsignificant reduction in FTI-treated HGPS cells at 24 and 48 hours adifferences between FTI-treated and nontreated unirradiated HGPSthose with normal and abnormal nuclear morphologies. “TFTI”=FTnuclear morphologies. (C) Examination of the average number of γafter FTI treatment, suggesting that the overall decrease in HGPS c“N”=cells with normal nuclear morphology. “NFTI”=FTI-treatednuclear morphology. “ABFTI”=FTI-treated cells with abnormal nufrom three experiments. For statistical analysis, cells frommultipleand a one-way ANOVA test was performed followed by a Tukey post-of means. Error bars represent the standard deviation (SD) of the nfrom cell to cell. Confocal images were captured using a Leica TCS

independent of nuclear morphology or may occur specifically incells whose nuclear morphology was normalized. The latterhypothesis is supported by our earlier results, which indicate acorrelation between abnormal nuclear morphology and impairedrepair of IR-induced DSBs. If there is no correlation betweenabnormal nuclear morphology and the DSB repair defect, wewould expect to see the following two outcomes: First, we wouldexpect to see an FTI-dependent reduction in the average number ofγ-H2AX foci per cell in HGPSAB fibroblasts due to individualHGPSAB fibroblasts that experience an improvement in DSBrepair but not in nuclear morphology. We did not observe this;FTI-treated HGPSAB fibroblasts exhibited the same averagenumber of γ-H2AX foci per cell as the nontreated HGPSABfibroblasts (Fig. 4C and Supplemental Fig. 3G), indicating thatimprovement in DSB repair only occurred in cells whose nuclearmorphology was normalized. Second, we would expect to see anFTI-dependent increase in the average number of γ-H2AXfoci per cell in HGPSN fibroblasts due to individual HGPSAB fibro-blasts that experience an improvement in nuclearmorphology, butnot in DSB repair. We did not observe this either; FTI-treatedHGPSN fibroblasts exhibited the same average number of γ-H2AXfoci per cell as nontreated HGPSN fibroblasts (Fig. 4C andSupplemental Fig. 3G), indicating that normalization of nuclearmorphology was always accompanied by an improvement in DSBrepair. These results demonstrate a strong correlation betweennuclear morphology normalization and improvement in repair ofIR-induced DSBs after FTI treatment.

MRN repair complex factor localization to IR-induced DSBsis delayed in HGPST cells and GFP-progerin-positive hCAECswith an abnormal nuclear morphology

Previous reports suggest that the DSB repair deficiency observed inHGPS cells is specific to HR repair [23]. It has been reported thatthe localization of the DSB repair factors, Rad50 and Rad51 to XPA-localized DSBs is impaired in HGPS cells due to the aberrantlocalization of XPA, possibly through steric hindrance [24].However, the localization of repair factors to IR-induced DSBshas not been clearly investigated in HGPS cells. Since the MRN

IR-induced DSBs in HGPS fibroblasts but has no effect on theals were treated for 72 hours with 5 μMFTI L-744832. Cells usedtil assayed. Immunocytochemistry was performed for γ-H2AX. (A) Percentage of unirradiated (white bars), 24 hours) fibroblasts exhibiting an abnormal nuclearmorphology in theleast 200 total cells were examined per group. Error barsthe average number of γ-H2AX foci per cell shows a statisticallyfter irradiation compared to non-treated HGPS cells. Thecells were not statistically significant. “T”=total cells includingI-treated total cells including those with normal and abnormal-H2AX foci shows no change in HGPSN and HGPSAB populationsells is due to correction of the nuclear morphology defect.cells with normal morphology. “AB”=cells with abnormalclear morphology. The data presented in B and C were obtainedexperiments were pooled together for each experimental group,test to obtain p values for the differences between specific pairsumber of γ-H2AX foci per cell to highlight the large varianceSP2 using Leica software. *p<0.05, **p<0.01, and ***p<0.001.


(MRE11/Rad50/phospho-NBS1) repair complex is essential forDSB HR repair (reviewed in Williams et al. [32]), we examined theprotein levels of NBS1 and MRE11 as well as their colocalizationkinetics to IR-induced γ-H2AX foci in total HGPS cells as well asGFP-progerin-positive hCAECs with an abnormal nuclear mor-phology (Fig. 5 and Supplemental Figs. 3H and I). Immunocyto-chemistry was used to examine colocalization, identified byoverlap of γ-H2AX shown in red and either phospho-NBS1 (Fig.5A) orMRE11 (Fig. 5B) shown in green. The percent of γ-H2AX foci

Fig. 5 – MRN repair complex factors exhibit delayed localization texpressing hCAECs with an abnormal nuclear morphology. Immuphospho-NBS1 orMRE11 at various time points after γ-irradiationMRE11 colocalization was determined. (A) Immunocytochemistryof γ-H2AX foci exhibiting phospho-NBS1 colocalization. (C) Immu(D) Percentage of γ-H2AX foci exhibiting MRE11 colocalization. (Eexperiments. Three experiments were performed. At least 1000 focells from multiple experiments were pooled together for each exfollowed by a Tukey post-test to obtain p values for the differencestandard deviation (SD) in the percentage of γ-H2AX foci with phocaptured using a Leica TCS SP2 using Leica software. Scale bar equ

that have either phospho-NBS1 (Fig. 5C and Supplemental Fig. 3H)or MRE11 (Fig. 5D and Supplemental Fig. 3I) colocalized at 0.5 and2 hours after irradiation was calculated. HGPS AG11513T (77%)and AG11498T (79%) fibroblasts and GFP-progerin expressinghCAECs with an abnormal nuclear morphology (75%) exhibitedstatistically significant lower percentages of γ-H2AX foci coloca-lized with phospho-NBS1 at 0.5 hours after irradiation comparedto wild-type AG08470T (93%) and AG16409T (92%) fibroblasts andwild-type hCAECs (92%) (Fig. 5C and Supplemental Fig. 3H).

o IR-induced DSBs in total HGPS fibroblasts and GFP-progerinnocytochemistry was performed for γ-H2AX and eitherand the percentage of γ-H2AX foci exhibiting phospho-NBS1 orfor γ-H2AX (green) and phospho-NBS1 (red). (B) Percentagenocytochemistry for γ-H2AX (green) and MRE11 (red).) Western analysis of total NBS1 and MRE11. Data are from twoci were examined for each time point. For statistical analysis,perimental group, and a one-way ANOVA test was performeds between specific pairs of means. Error bars represent thespho-NBS1 orMRE11 colocalized per cell. Confocal images wereals 5 μm. *p<0.05, **p<0.01, and ***p<0.001.

image of Fig.�5


Similarly, HGPS AG11513T (62%) and AG11498T (61%) fibroblastsand GFP-progerin expressing hCAECs with an abnormal nuclearmorphology (58%) exhibited a statistically significant lowerpercentage of γ-H2AX foci colocalized with MRE11 at 0.5 hoursafter irradiation compared to wild-type AG08470 (93%) andAG16409 (92%) fibroblasts and wild-type hCAECs (92%) (Fig. 5Dand Supplemental Fig. 3I). The differences in colocalizationbetween HGPST fibroblasts and wild-type fibroblasts as well asbetween GFP-progerin-positive hCAECs with abnormal nuclearmorphology and wild-type hCAECs disappeared by 2 hours afterirradiation for both phospho-NBS1 (Fig. 5C and SupplementalFig. 3H) and MRE11 (Fig. 5D and Supplemental Fig. 3I). Westernblot analysis showed no substantial differences in the proteinlevels of total NBS1 and MRE11 between wild-type and HGPSfibroblasts (Fig. 5E). Altogether, these results indicate thatphospho-NBS1 and MRE11 localization to IR-induced DSBs maybe delayed by as much as 2 hours. It is important to note that totalHGPS cells were examined so the localization defect is likely to beeven greater for HGPS cells with an abnormal nuclear morphology.Lastly, the observation that only select DSBs exhibit a lack ofcolocalization at 0.5 hours after irradiation suggests that the delayonly occurs at particular DSBs and is not a general phenomenonthroughout the entire nucleus.

Discussion

Recent reports have demonstrated an accumulation of DSBs inHGPS cells which may be due to defective DSB repair [19,23–25].These observations are interesting since DNA damage accumula-tion may contribute to the increased levels of apoptosis andsenescence observed in HGPS cells and in turn to the acceleratedaging phenotype. Here, we set forth to examine the severity of theDSB repair defect and investigate the mechanism/s responsible.Quantitative analysis showed that HGPS fibroblasts as well hCAECsexpressing a GFP-progerin construct exhibit a higher steady-statelevel of DSBs and impaired repair of IR-induced DSBs compared towild-type fibroblasts and wild-type hCAECs (Figs. 1B, 2B, and 3B).The higher steady-state level of DSBs may be responsible for theelevated DNA damage signaling pathways reported in HGPS cells[19]. Our results are in accordance with two recent studies whichshowed that HGPS cells exhibit a higher steady-state level of DSBs[24,25]. Sedelnikova et al. reported that the average steady-statenumber of γ-H2AX foci per cell increases with passage in HGPScultures. The authors detected 1.0, 1.2, and 1.9 average γ-H2AXfoci per cell at passage 15, 26, and 38, respectively. Liu et al. [24]also recently reported that passage 16 AG11513A HGPS primaryfibroblast cultures exhibited about 10 average γ-H2AX foci per cell.Our detection of 0.5 foci and 0.43 foci per cell for AG11513 andAG11498 HGPS cell cultures at passage 8–9 is in close agreementwith the results obtained by Sedelnikova et al. [25].

Separation of HGPS and GFP-progerin-positive cells into groupswith “normal” (N) or “abnormal” (AB) nuclear morphologiesshowed that the elevated steady-state level of DSBs and the DSBrepair defect are only present in cells with an abnormal nuclearmorphology (Figs. 1B, 2B, and 3B). The strong correlation betweenthe DSB associated defects and the abnormal nuclear morphologyhighlights the possibility of a mechanistic link even though it ispossible. If there is a mechanistic link, it is theoretically possiblethat the DSB-associated defects cause the abnormal nuclear

morphology. However, this scenario is unlikely since there is noreport to our knowledge suggesting that DSB formation leads tonuclear morphological changes similar to those observed in HGPS.Apoptosis, which can be caused by elevated levels of DSBs, doesresult in nuclear morphological changes, including chromatincondensation, nuclear shrinkage, and formation of apoptoticbodies, but these processes are characterized by a unique anddifferent morphological pattern (reviewed in Martelli et al. [33]).Alternatively, the nuclear structural irregularities that underlie theabnormal nuclear morphology may cause the DSB repair defect.This last possibility is intriguing since it is becoming clear that thenucleus is a highly ordered compartment and that nuclearstructure and chromatin organization are intricately linked tonuclear biological processes (reviewed in Folle [34] and Schneiderand Grosschedl [35]). Interestingly, the observation that wild-typecells with an abnormal nuclear morphology similar to that seen inHGPS cells also exhibit an elevated steady-state level of DSBscompared to wild-type fibroblasts with a normal nuclear mor-phology strengthens the possibility that nuclear architecturaldefects affect DSB levels. Further, this observation supports thehypothesis that the DSB-associated defects in HGPS cells arecaused by progerin-induced nuclear structural irregularities.

How can nuclear structural irregularities impair DSB repair?Our observations (Fig. 5) along with those of others [24], whichshow that DSB repair factor localization to DSBs is compromised inHGPS cells although protein levels are normal, highlight thepossibility that abnormal nuclear structure impairs DSB repair byphysically interfering with repair factor localization. Such physicalinterference could occur in multiple ways. First, the nuclear lobesand invaginations present in the nuclear envelope may compart-mentalize the nucleus and thus restrict the motility or accessibilityof repair factors to particular DSBs. A second possibility is that thelocal nuclear matrix, which may be necessary for the formation offunctional multiprotein repair complexes and to which laminshave been proposed to contribute [6], is disrupted by the presenceof progerin. Third, local chromatin organization, which appears tobe altered in HGPS cells [8,15], may interfere with the formation offunctional repair complexes. These scenarios, either alone or incombination, could render local regions of the nucleus “unfavor-able” to DSB repair, and DSBs that randomly occur in such“unfavorable” regions may experience delayed repair. Thishypothesis is supported by the our observation that completerepair of individual IR-induced DSBs in both wild-type and HGPScells occurs over a temporal continuum that appears to beginwithin 2 hours of irradiation and that may last beyond 24 hours,suggesting that the specific local nucleotide sequence and/or thelocal nuclear architecture can influence DSB repair significantly. Itis important to note that impaired localization of MRN repairfactors to DSBs may not just result in delayed repair, but precluderepair altogether. It has been recently proposed that among themultiple roles the MRN complex has in DSB repair is a tetheringfunction that acts to hold the two ends of the broken chromosometogether (reviewed in [32]). It is conceivable that a delay in theformation of such a tethering unit due to impaired localization ofMRN complex factors may lead to separation of the broken ends ata DSB, resulting in irreparable DSBs.

How can the accumulation of DSBs lead to the accelerated agingphenotype observed at the organismal level in HGPS? First, thepersistence of individual DSBs due to delayed repair or thepresence of irreparable DSBs can activate DNA damage response


pathways whose activity levels have been shown to be elevated inHGPS cells [19,36]. Hyperactive DNA damage response pathwayscan lead to higher rates of apoptosis and senescence which havealso been reported in HGPS cells. Even a delay of 1–2 hours asobserved for phospho-NBS1 and MRE11 localization may besufficient to lead to apoptosis or senescence. The importance ofearly response to DNA damage for cell survival is highlighted by arecent study, which showed that the radiosensitization caused by aone hour delay in activation of ATM after γ-irradiation accounts for70% of the radiosensitization caused by a 17-hour delay [37].Finally, higher levels of apoptosis and senescence can lead toaccelerated tissue/organ degeneration over time perhaps in partthrough exhaustion of adult stem cell pools as has been proposed(reviewed in Halaschek-Wiener et al. [38]).

We also demonstrate that FTI treatment, which has beenshown to improve multiple HGPS phenotypes [16–22], alsoimproves repair of IR-induced DSBs in HGPS cells. Interestingly,although we observed a reduction in the steady-state averagenumber of DSBs per cell (non-IR-induced), it was not statisticallysignificant, indicating that FTI treatment may not be able to reducethe level of steady-state DSBs. This observation is consistent with aprevious report, which showed that FTI treatment did not reducethe elevated levels of DNA damage signaling pathways in HGPScells [19]. There are several possible explanations for thediscrepancy in the effect of FTI treatment on the repair of IR-induced DSBs and steady-state DSB levels. One possibility is thatirradiated cells underwent longer FTI treatment (as described inthe Materials and methods) which may have produced a greaterbeneficial effect on DSB repair. Another possible explanation is thatFTI treatment creates a nuclear environment in which novel-DSBs(IR-induced) undergo normal repair, but preexisting DSBs thatformed in the context of nuclear structural irregularities remainirreparable. Lastly, there is evidence to suggest that someendogenous DSBs in HGPS cells (those observed at steady state)appear to have XPA aberrantly localized and may be functionallydifferent in origin and/or repair from those induced ectopically[24]. In this scenario, it is possible that FTIs improve the repair ofectopically induced DSBs but that the presence of XPA or othercharacteristics of particular endogenous DSBs detected in nonir-radiated cells may resist the FTI-based improvement. Furtherinvestigation is necessary to elucidate this matter. Such investiga-tion is essential because if FTIs do not improve endogenouslyoccurring DSBs in HGPS cells, then FTI treatment may not havesignificant clinical benefit.

HGPS research is motivated by the objective to developtherapeutic approaches for individuals with HGPS as well as bythe rationale that the mechanism/s underlying the acceleratedaging phenotype will increase our understanding of normal agingand age-associated disease. Regarding the former, our findingsoffer mixed results. On the one hand, we show that FTI treatmentcan improve the DSB repair of radiologically induced DSBs in HGPScells. On the other hand, our results and those of others suggestthat a subset of endogenously occurring DSBs (XPA-localized) maybe functionally distinct and resistant to FTI-based repair improve-ment. In terms of HGPS research expanding our understanding ofnormal aging, our results suggest that, similar to other progeroidsyndromes, HGPS pathology occurs due to an amplification of abiological process already believed to contribute to normal aging,namely DNA damage accumulation. The possibility that DNAdamage accumulation is amplified by nuclear structural irregular-

ities, as identified by our results, suggests that maintenance ofproper nuclear structural organization is important for longevity.

Acknowledgments

We would like to thank Dr. Ahmi Ben-Yehudah, Sandra Tavares-Varum, and Olga Momcilovic for helpful discussion. This researchwas supported by a grant from the National Institute of ChildHealth and Human Development, 1PO1HD047675.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.yexcr.2010.05.015.

R E F E R E N C E S

[1] R.L. Pollex, R.A. Hegele, Hutchinson-Gilford progeria syndrome,Clin. Genet. 66 (2004) 375–381.

[2] M. Eriksson, W.T. Brown, L.B. Gordon, M.W. Glynn, J. Singer, L.Scott, M.R. Erdos, C.M. Robbins, T.Y. Moses, P. Berglund, A. Dutra,E. Pak, S. Durkin, A.B. Csoka, M. Boehnke, T.W. Glover, F.S. Collins,Recurrent de novo point mutations in lamin A causeHutchinson-Gilford progeria syndrome, Nature 423 (2003)293–298.

[3] A. De Sandre-Giovannoli, R. Bernard, P. Cau, C. Navarro, J. Amiel, I.Boccaccio, S. Lyonnet, C.L. Stewart, A. Munnich, M. Le Merrer, N.Levy, Lamin a truncation in Hutchinson-Gilford progeria, Science300 (2003) 2055.

[4] D.W. Fawcett, On the occurrence of a fibrous lamina on the inneraspect of the nuclear envelope in certain cells of vertebrates, Am.J. Anat. 119 (1966) 129–145.

[5] U. Aebi, J. Cohn, L. Buhle, L. Gerace, The nuclear lamina is ameshwork of intermediate-type filaments, Nature 323 (1986)560–564.

[6] P. Hozak, A.M. Sasseville, Y. Raymond, P.R. Cook, Lamin proteinsform an internal nucleoskeleton as well as a peripheral lamina inhuman cells, J. Cell Sci. 108 (Pt 2) (1995) 635–644.

[7] T. Dechat, K. Pfleghaar, K. Sengupta, T. Shimi, D.K. Shumaker, L.Solimando, R.D. Goldman, Nuclear lamins: major factors in thestructural organization and function of the nucleus and chromatin,Genes Dev. 22 (2008) 832–853.

[8] R.D. Goldman, D.K. Shumaker, M.R. Erdos, M. Eriksson, A.E.Goldman, L.B. Gordon, Y. Gruenbaum, S. Khuon, M. Mendez, R.Varga, F.S. Collins, Accumulation of mutant lamin A causesprogressive changes in nuclear architecture in Hutchinson-Gilfordprogeria syndrome, Proc. Natl Acad. Sci. USA 101 (2004)8963–8968.

[9] K.N. Dahl, P. Scaffidi, M.F. Islam, A.G. Yodh, K.L. Wilson, T. Misteli,Distinct structural and mechanical properties of the nuclearlamina in Hutchinson-Gilford progeria syndrome, Proc. Natl Acad.Sci. USA 103 (2006) 10271–10276.

[10] V.L. Verstraeten, J.Y. Ji, K.S. Cummings, R.T. Lee, J. Lammerding,Increased mechanosensitivity and nuclear stiffness inHutchinson-Gilford progeria cells: effects of farnesyltransferaseinhibitors, Aging Cell 7 (2008) 383–393.

[11] K. Cao, B.C. Capell, M.R. Erdos, K. Djabali, F.S. Collins, A lamin Aprotein isoform overexpressed in Hutchinson-Gilford progeriasyndrome interferes with mitosis in progeria and normal cells,Proc. Natl Acad. Sci. USA 104 (2007) 4949–4954.

[12] T. Dechat, T. Shimi, S.A. Adam, A.E. Rusinol, D.A. Andres, H.P.Spielmann, M.S. Sinensky, R.D. Goldman, Alterations in mitosis



and cell cycle progression caused by a mutant lamin A known toaccelerate human aging, Proc. Natl Acad. Sci. USA 104 (2007)4955–4960.

[13] M. Columbaro, C. Capanni, E. Mattioli, G. Novelli, V.K. Parnaik, S.Squarzoni, N.M. Maraldi, G. Lattanzi, Rescue of heterochromatinorganization in Hutchinson-Gilford progeria by drug treatment,Cell. Mol. Life Sci. 62 (2005) 2669–2678.

[14] P. Scaffidi, T. Misteli, Reversal of the cellular phenotype in thepremature aging disease Hutchinson-Gilford progeria syndrome,Nat. Med. 11 (2005) 440–445.

[15] D.K. Shumaker, T. Dechat, A. Kohlmaier, S.A. Adam,M.R. Bozovsky,M.R. Erdos, M. Eriksson, A.E. Goldman, S. Khuon, F.S. Collins, T.Jenuwein, R.D. Goldman, Mutant nuclear lamin A leads toprogressive alterations of epigenetic control in premature aging,Proc. Natl Acad. Sci. USA 103 (2006) 8703–8708.

[16] M.W. Glynn, T.W. Glover, Incomplete processing of mutant laminA in Hutchinson-Gilford progeria leads to nuclear abnormalities,which are reversed by farnesyltransferase inhibition, Hum. Mol.Genet. 14 (2005) 2959–2969.

[17] J.I. Toth, S.H. Yang, X. Qiao, A.P. Beigneux, M.H. Gelb, C.L. Moulson,J.H. Miner, S.G. Young, L.G. Fong, Blocking proteinfarnesyltransferase improves nuclear shape in fibroblasts fromhumans with progeroid syndromes, Proc. Natl Acad. Sci. USA102 (2005) 12873–12878.

[18] L.G. Fong, D. Frost, M. Meta, X. Qiao, S.H. Yang, C. Coffinier, S.G.Young, A protein farnesyltransferase inhibitor amelioratesdisease in a mouse model of progeria, Science 311 (2006)1621–1623.

[19] Y. Liu, A. Rusinol, M. Sinensky, Y. Wang, Y. Zou, DNA damageresponses in progeroid syndromes arise from defective maturationof prelamin A, J. Cell Sci. 119 (2006) 4644–4649.

[20] S.H. Yang, M. Meta, X. Qiao, D. Frost, J. Bauch, C. Coffinier, S.Majumdar, M.O. Bergo, S.G. Young, L.G. Fong, Afarnesyltransferase inhibitor improves disease phenotypes inmice with a Hutchinson-Gilford progeria syndrome mutation,J. Clin. Invest. 116 (2006) 2115–2121.

[21] S.H. Yang, X. Qiao, L.G. Fong, S.G. Young, Treatment with afarnesyltransferase inhibitor improves survival in mice with aHutchinson-Gilford progeria syndrome mutation, Biochim.Biophys. Acta 1781 (2008) 36–39.

[22] B.C. Capell, M.R. Erdos, J.P. Madigan, J.J. Fiordalisi, R. Varga, K.N.Conneely, L.B. Gordon, C.J. Der, A.D. Cox, F.S. Collins, Inhibitingfarnesylation of progerin prevents the characteristic nuclearblebbing of Hutchinson-Gilford progeria syndrome, Proc. NatlAcad. Sci. USA 102 (2005) 12879–12884.

[23] B. Liu, J. Wang, K.M. Chan, W.M. Tjia, W. Deng, X. Guan, J.D.Huang, K.M. Li, P.Y. Chau, D.J. Chen, D. Pei, A.M. Pendas, J.Cadinanos, C. Lopez-Otin, H.F. Tse, C. Hutchison, J. Chen, Y. Cao,K.S. Cheah, K. Tryggvason, Z. Zhou, Genomic instability inlaminopathy-based premature aging, Nat. Med. 11 (2005)780–785.

[24] Y. Liu, Y. Wang, A.E. Rusinol, M.S. Sinensky, J. Liu, S.M. Shell, Y.Zou, Involvement of xeroderma pigmentosum group A (XPA) in

progeria arising from defective maturation of prelamin A, FASEB J.22 (2008) 603–611.

[25] O.A. Sedelnikova, I. Horikawa, C. Redon, A. Nakamura, D.B.Zimonjic, N.C. Popescu, W.M. Bonner, Delayed kinetics of DNAdouble-strand break processing in normal and pathological aging,Aging Cell 7 (2008) 89–100.

[26] C. Lois, E.J. Hong, S. Pease, E.J. Brown, D. Baltimore, Germlinetransmission and tissue-specific expression of transgenes deliveredby lentiviral vectors, Science 295 (2002) 868–872.

[27] E.P. Rogakou, D.R. Pilch, A.H. Orr, V.S. Ivanova, W.M. Bonner, DNAdouble-stranded breaks induce histone H2AX phosphorylation onserine 139, J. Biol. Chem. 273 (1998) 5858–5868.

[28] E.P. Rogakou, C. Boon, C. Redon, W.M. Bonner, Megabasechromatin domains involved in DNA double-strand breaks invivo, J. Cell Biol. 146 (1999) 905–916.

[29] F. Antonelli, M. Belli, G. Cuttone, V. Dini, G. Esposito, G. Simone, E.Sorrentino, M.A. Tabocchini, Induction and repair of DNAdouble-strand breaks in human cells: dephosphorylation ofhistone H2AX and its inhibition by calyculin A, Radiat. Res. 164(2005) 514–517.

[30] I.B. Nazarov, A.N. Smirnova, R.I. Krutilina, M.P. Svetlova, L.V.Solovjeva, A.A. Nikiforov, S.L. Oei, I.A. Zalenskaya, P.M. Yau, E.M.Bradbury, N.V. Tomilin, Dephosphorylation of histonegamma-H2AX during repair of DNA double-strand breaks inmammalian cells and its inhibition by calyculin A, Radiat. Res. 160(2003) 309–317.

[31] R. Shroff, A. Arbel-Eden, D. Pilch, G. Ira, W.M. Bonner, J.H. Petrini,J.E. Haber, M. Lichten, Distribution and dynamics of chromatinmodification induced by a defined DNA double-strand break,Curr. Biol. 14 (2004) 1703–1711.

[32] R.S. Williams, J.S. Williams, J.A. Tainer, Mre11–Rad50–Nbs1 is akeystone complex connecting DNA repair machinery, double-strand break signaling, and the chromatin template, Biochem. CellBiol. 85 (2007) 509–520.

[33] A.M. Martelli, M. Zweyer, R.L. Ochs, P.L. Tazzari, G. Tabellini, P.Narducci, R. Bortul, Nuclear apoptotic changes: an overview, J.Cell. Biochem. 82 (2001) 634–646.

[34] G.A. Folle, Nuclear architecture, chromosome domains andgenetic damage, Mutat. Res. 658 (2008) 172–183.

[35] R. Schneider, R. Grosschedl, Dynamics and interplay of nucleararchitecture, genome organization, and gene expression, GenesDev. 21 (2007) 3027–3043.

[36] I. Varela, J. Cadinanos, A.M. Pendas, A. Gutierrez-Fernandez, A.R.Folgueras, L.M. Sanchez, Z. Zhou, F.J. Rodriguez, C.L. Stewart, J.A.Vega, K. Tryggvason, J.M. Freije, C. Lopez-Otin, Accelerated ageingin mice deficient in Zmpste24 protease is linked to p53 signallingactivation, Nature 437 (2005) 564–568.

[37] J.S. White, S. Choi, C.J. Bakkenist, Irreversible chromosomedamage accumulates rapidly in the absence of ATM kinaseactivity, Cell Cycle 7 (2008).

[38] J. Halaschek-Wiener, A. Brooks-Wilson, Progeria of stem cells:stem cell exhaustion in Hutchinson-Gilford progeria syndrome,J. Gerontol. A Biol. Sci. Med. Sci. 62 (2007) 3–8.

Documents

Defective DSB repair correlates with abnormal nuclear morphology and is improved with FTI treatment in Hutchinson-Gilford progeria syndrome fibroblasts