Constitutive Activation of Caspase-3 and Poly ADP Ribose Polymerase Cleavage in Fanconi Anemia Cells

Published OnlineFirst January 12, 2010; DOI: 10.1158/1541-7786.MCR-09-0373

Molecular DNA Damage and Cellular Stress Responses Cancer Research Constitutive Activation of Caspase-3 and Poly ADP Ribose Polymerase Cleavage in Fanconi Anemia Cells

Alex Lyakhovich1,2 and Jordi Surrallés1,3

Abstract Fanconi anemia (FA) is a rare syndrome characterized by developmental abnormalities, progressive bone marrow failure, and cancer predisposition. Cells from FA patients exhibit hypersensitivity to DNA cross-linking agents and oxidative stress that may trigger apoptosis. Damage-induced activation of caspases and poly ADP ribose polymerase (PARP) enzymes have been described for some of the FA complementation groups. Here, we show the constitutive activation of caspase-3 and PARP cleavage in the FA cells without exposure to exogenous DNA-damaging factors. These effects can be reversed in the presence of reactive oxygen species scavenger N- acetylcystein. We also show the accumulation of oxidized proteins in FA cells, which is accompanied by the tumor necrosis factor (TNF)-α oversecretion and the upregulation of early stress response kinases pERK1/2 and p-P38. Suppression of TNF-α secretion by the extracellular signal-regulated kinase inhibitor PD98059 results in reduction of caspase-3 cleavage, suggesting a possible mechanism of caspases-3 activation in FA cells. Thus, the current study is the first evidence demonstrating the damage-independent activation of caspase-3 and PARP in FA cells, which seems to occur through mitogen-activated protein kinase activation and TNF-α oversecretion. Mol Cancer Res; 8(1); 46–56. ©2010 AACR.

Introduction Cells derived from FA patients are hypersensitive to cross- linking agents that are potent inducers of programmed cell Fanconi anemia (FA) is a genetic disorder characterized death (PCD or apoptosis; ref. 6). Studies in cohorts of FA by congenital abnormalities, bone marrow failure, and patients show that PCD is uncommon in the absence of marked cancer susceptibility (1, 2). Currently, 13 FA com- damaging factors and the major cause of mortality is bone plementation groups are known to exist, each of which re- marrow failure (7-9). This can be mediated by several indu- presents a FA subtype abbreviated as A (FANCA), B cers, including dissolved oxygen, superoxide and hydroxyl (FANCB), C (FANCC), D1 (FANCD1/BRCA2), D2 radicals, and other reactive oxygen species (ROS), and re- (FANCD2), E (FANCE), F (FANCF), G (FANCG), I sults in cell growth arrest or cell death (10-12). FA cells (FANCI/KIAA1794), J (FANCJ/BRIP1), L (FANCL), M are sensitive to ROS, thus acquiring abnormalities in several (FANCM), and N (FANCN; refs. 3, 4). With the exception redox status end points (13-15). FANCA and FANCG are of FANCD2 and FANCI, the encoded FA proteins have redox-sensitive proteins that are multimerized to form a nu- no similarities with each other but cooperate in a common clear complex in response to oxidative stress (16). Not sur- FA/BRCA pathway by the formation of subnuclear and prisingly, the survival of human primary FA BM cells can be nuclear complexes, where activation by monoubiquitinyla- ameliorated by reducing oxidative stress (14). tion of FANCD2 and FANCI seems to orchestrate a cas- Hypersensitivity to ROS often acts as a second messenger cade of events in response to DNA damage (5). in cellular signaling and is mediated by specific cysteine pro- teases, known as caspases (17). Sequential activation of caspases, on induction of PCD, has been shown in FA Authors' Affiliations: 1Department of Genetics and Microbiology, hematopoietic progenitor cells in response to mutagenic Universitat Autònoma de Barcelona, Barcelona, Spain; 2Chris Doppler stress (18). Caspases exist within the cell as zymogens (inac- Laboratory on Molecular Cancer Chemoprevention, Medical University tive proforms) that can be cleaved to form active enzymes of Vienna, Vienna, Austria; and 3Centre for Biomedical Research on Rare Diseases, Bellaterra, Spain following the induction of apoptosis (19). One of the first Note: Supplementary data for this article are available at Molecular Can- proteins identified as a substrate for caspases is poly ADP cer Research Online (http://mcr.aacrjournals.org/). ribose (PAR) polymerase (PARP; also known as PARP1 Corresponding Author: Jordi Surrallés, Department of Genetics and Mi- and adenosine diphosphate ribosyltransferase; ref. 20). crobiology, Universitat Autònoma de Barcelona, 08193 Bellaterra (Barce- PARP is involved in DNA damage repair and catalyzes the lona), Barcelona, Spain. Phone: 34-93-581-1830; Fax: 34-93-581-2387. synthesis and binding of PAR to DNA strand breaks and E-mail: [email protected] and Alex Lyakhovich, Christian Doppler Laboratory on Molecular Cancer Chemoprevention, 1090 Vienna, Austria. modifying nuclear proteins (21). The ability of PARP to re- Phone: 43-1-40400-6154; Fax: 43-1-40400-4724. E-mail: alex.lyakho- pair damaged DNA is prevented following cleavage of [email protected] PARP by caspase-3 (21). doi: 10.1158/1541-7786.MCR-09-0373 Present research is aimed at studying caspase-3 activity as ©2010 American Association for Cancer Research. well as PARP status in FA cells of several complementation

46 Mol Cancer Res; 8(1) January 2010

Caspase-3/PARP Activation in Fanconi

groups in the absence of exogenous DNA damage. We have dissolved in DMSO, was used at 50 μmol/L final concentra- shown higher caspase-3 activity and PARP cleavage in FA- tion. ROS inhibitor NAC, from Sigma, was used at 2 mmol/L D2 and other FA complementation groups compared with final concentration. WP9QY (Abgent, Inc.), an inhibitor of their corrected counterparts or in the small interfering TNF-receptors, was used at 1 μmol/L concentration. RNA (siRNA) FANCD2–depleted normal cells versus scRNA-transfected cells. Interestingly, the level of PARP Cells Treatment and Preparation of Cell Lysates cleavage and caspase-3 upregulation could be reduced on For siRNA experiments, we followed previously de- treatment of cells with ROS inhibitor, suggesting that oxi- scribed procedures with the same FANCD2 siRNA se- dative stress is one of the mediators in these events. We also quences (Invitrogen; ref. 22). Whenever necessary, the showed higher level of apoptosis in FA cells with PKC inhibitor bisindolylmaleimide I (GF109203X from corresponding accumulation of oxidized proteins. This Calbiochem) was added directly to the cell medium as was accompanied by tumor necrosis factor (TNF)-α over- DMSO-dissolved stock solution. Cells were washed with secretion and upregulation of early damage-response ki- ice-cold PBS and resuspended in buffer C consisting of nases p-ERK1/2 and p-P38. Moreover, suppression of 20 mmol/L HEPES (pH 7.9), 420 mmol/L KCl, 25% α TNF- secretion by extracellular signal-regulated kinase glycerol, 0.1 mmol/L EDTA, 5 mmol/L MgCl2,0.2% (ERK) inhibitor PD98059 results in the reduction of cas- Nonidet P-40, 1 mmol/L DTT, and a 1:40 volume of pro- pase-3 cleavage, suggesting a possible mechanism of cas- tease inhibitor mixture (Sigma) on ice for 30 min. After pases-3 activation in FA cells. The blocking of TNF centrifugation at 12,000 × g for 10 min at 4°C, the super- receptor decreased caspase-3 activity in D2−/−,butnot natant was collected as the total cell lysate. The pellet was in D2−/−–corrected cells, suggesting that caspase-3 re- resuspended in buffer A [10 mmol/L HEPES (pH 7.9), α sponse is TNF /TNFR1 mediated. The intracellular 10 mmol/L KCl, 1.5 mmol/L MgCl2,20%glycerol, ROS seemed to be TNFα/TNFR1-dependent for D2−/− 1 mmol/L DTT, and a 1:40 volume of protease inhibitor but not for D2−/−–corrected cells, suggesting that overpro- mixture containing 4-(2-aminoethyl)benzenesulfonyl fluo- duction of TNFα, rather than ROS accumulation, causes ride, pepstatin A, E-64, bestatin, leupeptin, aprotinin, and caspase-3/PARP response. Thus, the current study is the sodium EDTA] and homogenized on ice by sonication. first evidence demonstrating the damage-independent acti- After centrifugation at 1,000 × g for 10 min at 4 C, the vation of caspase-3 and PARP in FA cells, and that this ac- supernatant was collected as the cytoplasmic fraction. The tivation probably occurs through mitogen-activated protein resulting nuclear pellet was then washed with buffer A twice kinases (MAPK) activation and TNF-α oversecretion. and resuspended in TNE buffer [10 mmol/L Tris-HCl (pH 7.5), 0.1 mol/L NaCl, 1 mmol/L EDTA]. For de- phosphorylation experiments, 100 units of λ-phosphatase Materials and Methods (Calbiochem) were added to the appropriate cell lysates following incubation for 30 min at 37°C. For the degradation Cell Lines and Culturing assay (Supplementary Fig. S1), siRNA FANCD2–depleted The following human transformed fibroblast cell lines samples were incubated in parallel with the mixture of were used in this study: PD20 (FANCD2−/−), PD20 retro- sodium vanadate and sodium pyrophosphate (both from virally corrected with pMMp-FANCD2 cDNA, FANCA−/ Sigma). To test for IFN response, we analyzed siRNA- − (PD220)–transformed fibroblasts and corrected counter- transfected primary cells or cells stimulated with 2 μg parts (FA consortium),4 FANC-G– and FANC-G– Poly-I-poly-C were followed by Western blot analyses of corrected primary fibroblasts (gift of Dr. D. Schindler, eIF2α/P-eIF2α antibodies or real-time PCR experiments Wurzburg University, Wurzburg, Germany). In addition, (22) to test the expression of the IFN pathway–related we used primary human fibroblasts derived from a healthy gene STAT1. individual and lymphoblastoid cells from a non-FA patient [FANCL (EUFA 868), FANCJ (LFA 177), and FANCA SDS-PAGE and Western Blot Analysis (LFA 82)–transformed lymphoblastoid cells]. FANCC Cell lysates (25 μg/lane) or purified proteins were re- (PD331.1)–transformed fibroblasts and corrected counter- solved on 4% to 8% or 4% to 13% SDS-polyacrylamide parts were cultured at 70% confluence as described previ- gels under reducing conditions (10 mmol/L DTT or 1% ously (22). FANCI and FANCM human lymphoblasts β-mercaptoethanol). Proteins were then transferred to a poly- were kind gifts from Juan Bueren's laboratory (Centro de vinylidene difluoride membrane and were probed with an Investigaciones Energéticas, Medioambientales y Tecnológi- anti-PARP, cleaved anti–caspase-3, cleaved anti–caspase-8, cas, Madrid, Spain). The PARP inhibitor 3-aminobenza- anti–glyceraldehyde-3-phosphate dehydrogenase, anti- mide (3AB; Sigma) was dissolved in DMSO and was vinclulin (all from AbCAM), anti-PAR antibody (a rabbit added to cell medium at a final concentration of 10 mmol/L polyclonal IgG from AbNOVA or SantaCruz), or with an when required. ERK inhibitor PD98059 (Calbiochem), anti-FANCD2 antibody (monoclonal mouse IgG, Novus Biologicals) followed by a horseradish peroxidase–conjugated secondary antibody (Amersham Biosciences). Proteins were visualized by using the enhanced chemiluminescence system 4 http://www.fanconi.org (Amersham Biosciences).

www.aacrjournals.org Mol Cancer Res; 8(1) January 2010 47

Lyakhovich and Surrallés

Immunoprecipitation were then scanned with a Typhoon Laser Scanner at 520 Cells were washed with PBS and resuspended in 1 mL of nm to identify signals of incorporated fluorescein-12 lysis buffer [50 mmol/L HEPES (pH 7.0), 100 mmol/L dUTP (apoptotic cells) or with 620 nm for identification NaCl,1mmol/LEDTA,and0.1%NP40]followedby of total cell numbers from propidium iodide staining. The shearing with a 23.5-gauge needle. Clarified supernatants apoptotic index was counted by 2D-Bio-Rad software as a (2 mg) were mixed with 1 μg of affinity-purified antisera result of three independent measurements. To confirm the against PARP overnight at 4°C. Immunocomplexes were above data, several cell samples were assessed for apoptosis precipitated with protein-A-Sepharose beads (Pierce), using an Annexin V–fluorescence-activated cell sorting washed thrice with lysis buffer, and boiled in 1× Laemmli protocol, and for staining using an Annexin V and 7-ami- buffer. All precipitated proteins were resolved by 4% to noactinomycin D with a BD ApoAlert Annexin V kit (BD 12% SDS-PAGE and transferred to Immobilon-P mem- Pharmingen) in accordance with the manufacturer's in- branes (Millipore), followed by Western blot analysis with structions. Apoptosis was analyzed by quantification of An- phospho-PARP antibodies (MBL Laboratories). nexin V–positive cell population by flow cytometry.

Assay of Caspase-3 Activity OxyBlot Analyses After the incubation with or without caspase-3 inhibitor, To assess the formation of protein carbonyl groups, the cells were washed thrice in PBS and cell extracts were pre- OxyBlot protein oxidation detection kit (Chemicon Int., pared as above. Equal volume of STKM buffer [0.25 mol/L Integen) was used according to the manufacturer's detailed sucrose, 50 mmol/L Tris-HCl (pH 7.5), 25 mmol/L KCl, protocol. Briefly, total protein extracts were denatured by 5 mmol/L MgCl2, and 0.25% Triton X-100] was added mixing with equal volumes of 12% SDS and then with the and the lysates remained on ice for 15 min. Cell lysates derivatized by adding 2,4-dinitrophenylhydrazine solution were stored at −80°C until assay. Caspase-3 activity was or control for 15 min at room temperature. After neutral- colorimetrically measured using Ac-Asp-Glu-Val-Asp ization, samples were resolved on 4% to 12% Tris-glycine 4-methyl-coumaryl-7-amide as a substrate following the gels followed by Western blotting. Blots were probed with manufacturer's manual (Roche). The assay was done by primary antibody recognizing specific derivatized dinitro- incubating 20 μL of cell lysates with 178 μL of reaction phenyl moities of the proteins. Proteins that underwent buffer [100 mmol/L HEPES (pH 7.5), 20% (v:v) glycerol, oxidative modification (i.e., carbonyl group formation) 5 mmol/L DTT, and 0.5 mmol/L EDTA] and 2 μLof were identified as a band in the derivatized sample, but 10 mmol/L substrate at 37°C for 2 h. Release of 7-amino- not in the control. Levels of oxidatively modified proteins 4-methyl-coumarin from Ac-Asp-Glu-Val-Asp 4-methyl- were quantified and expressed as fold increase versus nor- coumaryl-7-amide by the enzyme reaction was measured mal controls through measurement by scanning densitom- using a spectrophotometer (Bio-Rad.). Relative protease etry with the TotaLab2.0 image program. activity was determined by comparing the levels of the control or normal cells with deficient or depleted counterparts, Assay for Glutathione Peroxidase, Catalase, and with the number of relative fluorescence units detected in Intracellular ROS controls set at 100%. Corresponding levels of caspase-3 ac- Glutathione peroxidase activity was determined spectro- tivity in the presence of caspase3 inhibitors were subtracted photometrically by monitoring the oxidation of NADPH at to exclude background signal. Validation of this method was 340 nm (23). The Cellular Glutathione Peroxidase Assay alternatively confirmed by chemiluminescence technique kit (Calbiohem) was used to achieve this goal. Briefly, sam- using the FluorAce Apopain Assay kit (Bio-Rad), according ples containing 0.2 μg of total proteins were dissolved in to the manufacturer's instructions. The fluorescence was 1:10 (v.v) assay buffer [1 mmol/L reduced glutathione, measured using the VersaFluorTM fluorimeter (Bio-Rad) 0.22 mmol/L tert-butyl hydroperoxide, and 0.4 UY/mL after 30 min of incubation with the Ac-DEVD-AFC fluoro- glutathione reductase (pH 7.6)]. Immediately before assay, genic caspase-3 substrate. samples were mixed with NADPH reagent in a 96-well mi- crotiter plate and the absorbances at 340 nm were measured Analysis of Apoptosis in FA Cells over 3-min intervals using a recording spectrophotometer. − Apoptosis was assessed using slightly modified protocol An extinction coefficient of 6.1 mm 1 was used for the cal- described in the Apoptosis Detection Manuel (Promega), culation. One unit is the amount needed to oxidize 1 nmol which uses the principle of terminal deoxynucleotidyl of NADPH per minute. Catalase activity was determined transferase–mediated dUTP nick end labeling to detect spectrophotometrically at 240 nm by the reduction of fragmented DNA in apoptotic cells. Briefly, cells were H2O2 as described previously (24). GPX activity was mea- grown on slides or prepared as cytospins and fixed in 3% sured by monitoring NADPH oxidation. A standard reac- paraphormaldehyde, were permeabilized, were washed, tion mixture (0.2 mL) containing 100 mmol/L Tris-HCl and were treated with the TdT incubation buffer at 37C (pH8.0),5mmol/LEDTA,0.2mmol/Lβ-NADPH, for 60 min. After thorough washing in PBS, the slides were 1 mmol/L NaN3, 3 mmol/L glutathione, 0.1% (v/v) Triton counterstained with 1 μg/mL propidium iodide (Sigma) in X-100, 1.4 units of glutathione reductase, and 2.5 μmol/L PBS (15 min) and were washed in distilled water before TcGPXII was incubated at 30°C for 5 min. The back- mounting in fluorescence-free mountant (DAKO). Slides ground rate of NADPH oxidation was determined and

48 Mol Cancer Res; 8(1) January 2010 Molecular Cancer Research

Caspase-3/PARP Activation in Fanconi

the reaction was initiated by the addition of the hydroper- DNA counter stain (Vector Laboratories). Microscopic oxide substrate. The nonenzymic activity due to the auto- analysis and quantification were done using an Olympus oxidation of glutathione was also examined. The enzyme BX-50 fluorescence microscope. Images were captured − − activity was calculated using e = 6220 M 1·cm 1. Evalua- with a Leica DMRB epifluorescence microscope equipped tion of intracellular ROS levels was estimated using the ox- with a DC 200 digital camera and the Leica DC Viewer idant-sensitive H2-DCFDA probe as described by Royall capturing software. Images were finally computer edited and Ischiropoulos [Royall, J.A. and Ischiropoulos, H. using Photodeluxe (Adobe version 2.0). (1993). Evaluation of 2′,7′-dichlorofluorescein and dihy- drorhodamine 123 as fluorescent probes for intracellular Statistical Analysis H2O2 in cultured endothelial cells. Arch. Biochem. Bio- Statistical analyses of data from four replicate trials were phys. 302, 348-355]. Cells were grown in 6-multiwell carried out by ANOVA and Fisher's protected least signif- plates, then incubated in serum-free (phenol red–free icant difference test using the STATVIEW program (Aba- RPMI 1640) medium with H2-DCFDA (5 μmol/L) cus Concepts, Inc.). All percentage values were subjected for 30 min as described above. Cells were then carefully to arcsine transformation before statistical analysis. All va- washed in PBS, scraped off into 1 mL of water, and the lues are expressed as mean ± SEM. A probability of P < fluorescence was measured (excitation at 495 nm and emis- 0.05 was considered to be statistically significant. sion at 520 nm). Results TNF-α Titration Overproduction of TNF-α in the supernatant of expo- PARP Cleavage in FA Cells nentially growing FA cells with or without NAC inhibitor Our proteomics data suggest that PARP family proteins, was done according to Quantikine ELISA as described by including caspase substrates are differentially expressed in the supplier (R&D Systems). Briefly, cells (either scrapped FA cells and associated the FANCD2 immunoprecipitated by rubber policemen or in suspension) were collected by pellets (Table 1).5 Because PARP plays a role in apoptosis, centrifugation and supernatants were filtered through a we studied the PARP level in various FA complementation 0.45-μm Millipore Filter (Millipore) and were immediately groups. We reinvestigated the previously reported lack of used for assay or stored at −80°C. Samples were diluted in PARP activity in fibroblasts from FA patients (25) by de- appropriate buffer and were added to the anti–TNF-α-an- termining the level of PARP in the cells from a FA-D2 tibody precoated microwell strips. After washing away any complementation group that usually represents a more se- unbound substances, an enzyme-linked polyclonal anti- vere phenotype compared with upstream (core) FA sub- body specific for TNF-α was added to the wells. Following groups (26). Our Western blot data suggest that the incubation and washing steps, a substrate solution was PARP is cleaved in FA-D2–deficient (PD20) but not in added to the wells and color developed in proportion to the corrected (PD20cor) cells. This event seems to be in- the amount of TNF-α bound in the initial step. After stop- dependent from FANCD2 monoubiquitinylation as only ping the reaction, the intensity of the color was measured moderate PARP cleavage was observed in a nonmonoubi- by microplate reader at 450 nm with the correction wave- quitinable mutant (K561R) counterpart (Fig. 1A). Pre- length set at 540 nm. treatment of cells with PARP inhibitor, 3AB, led to reduction of PARP cleavage (Fig. 1B). To further examine Immunofluorescence whether PARP cleavage can occur in normal cells con- Cells grown on coverslips were washed with PBS and verted by siRNA to FA-phenotype, we performed subsequently fixed by adding PBS containing 4% formal- FANCD2 knockdown experiments by transfecting prima- dehyde (Sigma-Aldrich) for 10 min at room temperature. ry human fibroblasts with FANCD2 siRNA. We observed Cells were washed with PBS and incubated with PBS and the appearance of cleaved forms of PARP 1 day after trans- 0.5% Triton (Sigma-Aldrich) for 15 min, subsequently fection (Fig. 1C). Interestingly, the upper band rinsed with a washing buffer (WB) consisting of 3% corresponding to the uncleaved 116-kDa PARP seemed bovine albumin (Sigma-Aldrich) and 0.05% Tween 20 to be 8 to 10 kDa higher in the FANCD2-depleted sam- (Sigma-Aldrich) in PBS, incubated with corresponding ples, suggesting possible PARP modification. Because primary antibody in WB overnight at 4°C, and washed PARP can be activated by phosphorylation (27), we tested for 15 min in WB with gentle agitation. In addition to such possibility by treatment of the cell lysates with the antibodies used for Western blotting, cleaved PARP λ-phosphatase. Reversal of the mobility shift suggested PARP (Asp214) antibody (human specific) #9541 (Cell Signal- phosphorylation on depletion of FANCD2 (Fig. 1C). Sim- ing, Inc.) has been used. Incubation with secondary anti- ilar reversal was obtained by the pretreatment of FANCD2 mouse Alexa Fluor 488 and anti-rabbit Alexa Fluor 568 siRNA–depleted cells with GF109203X, the inhibitor of antibodies (both from Molecular Probes) diluted in WB PKC that evokes PARP phosphorylation in vitro.This was done at 4°C for 80 min followed by a 15-min washing led to the decrease in intensity of the upper PARP band step in WB with gentle agitation. After the last antibody labeling step, the preparations were mounted in anti-fading medium containing 4′-6′-diamidino-2-phenylindole as a 5 Our unpublished data.

www.aacrjournals.org Mol Cancer Res; 8(1) January 2010 49

Lyakhovich and Surrallés

Table 1. Caspase-related proteins co-immunoprecipitated with FANCD2 protein

Protein name* Protein ID Description

PACAP A5A3E0 Proapoptotic caspase adaptor protein that interacts and inhibits caspase-3 activity PKN1 NP_002732.3 PKN1 is a fatty acid and Rho-activated serine/threonine protein kinase Interacts with TNFα and proteolytically activated by caspase-3 MAP1A NP_002364.5 Microtuble-associated protein 1A, in which proteolysis results from the Aβ-mediated activation of caspase-3 and calpain Calpastatin NP_001139540.1 Calpastatin-type II is a highly specific inhibitor of calpain, in which activation is essential for the enzymatic activation of the critical effector caspase-3 NAP1L3 NP_004529.2 Nucleosome assembly protein 1-like 3 is a substrate for caspase-3 CARD10 (CARMA3) ENSG00000100065 Caspase recruitment domain family, member 10

*Proteins identified by matrix-assisted laser desorption/ionization–time-of-flight technique in immunopellets of FANCD2-immunoprecipitated cell extracts.

(Fig. 1C, right). To confirm this event, we immunoprecipi- PARP in the FANCD2-depleted samples and partial de- tated cell fractions after siRNA FANCD2 depletion with a crease of phosphor-PARP in the λ-phosphatase–treated PARP antibody followed by Western blotting and probing samples, whereas total PARP level remained unchanged with anti–phospho-specific PARP antibodies. We recog- (Fig. 1D). These data confirm that conversion of normal nized increased signals of multiple bands of phosphorylated cells into FA phenotype is accompanied by PARP cleavage

FIGURE 1. PARP cleavage and upregulation in FA cells. A. Immunoblot showing PARP cleavage in FANCD2-deficient (PD20), FANCD2-proficient (PD20cor), or nonmonoubiquitinable mutant (K561R) cells. B. Immunoblot showing PARP processing in PD20 and PD20-corrected cells treated (+) or untreated (−) with PARP inhibitor 3AB. C. PARP cleavage is increased in primary fibroblasts transfected with FANCD2 siRNA for the corresponding time intervals. Arrows, mobility shift representing PARP phosphorylation. Note the reduction of PARP signal in the sample treated with λ-phosphatase or in the siRNA FANCD2–depleted cells pretreated with PKC inhibitor. D. Immunoblot of the above FANCD2 knockdown samples immunoprecipitated with PARP antibody followed by probing with phosphor-PARP or total PARP antibodies. E. Immunoblots showing PARP cleavage in different FA subgroups treated or untreated with the PARP inhibitor, 3AB.

50 Mol Cancer Res; 8(1) January 2010 Molecular Cancer Research

Caspase-3/PARP Activation in Fanconi

FIGURE 2. FA cells show higher caspase-3 activity and apoptotic levels versus corrected counterparts. A. Immunoblot shows the levels of cleaved caspases-3 in FA cells treated or untreated with NAC. B. Immunofluorescent image shows the levels of cleaved caspase-3 (red channel) in PD20 (A and A*), K561R (B and B*), or PD20-corrected (C and C*) cells. C. Colorimetric measurement of caspase-3 activity in FA cells (P* < 0.05). D. Apoptotic index of FA cells as measured by terminal deoxynucleotidyl transferase–mediated dUTP nick end labeling–based assay. Example of Typhoon laser–scanned PD20cor cells stained either with propidium iodide to define apoptotic cells (scanned in red channel at 620 nm) or with fluorescein 12-dUTP to define total number of cells (green channel at 520 nm; left). The percentage of apoptosis was determined by comparing two scanned images using 2D-Bio-Rad software applied for three independent measurements (right). P* < 0.01.

and phosphorylation. Importantly, the appearance of PARP groups, we performed Western blot analyses of the cell ex- in the upper band was specific to phoaphorylated PARP as tracts isolated from FANCG fibroblasts, FANCL (EUFA pretreatment of cell samples with phosphatase inhibitors re- 868), FANCJ (LFA 177), and FANCA (LFA 82)–trans- stored the appearance of the upper band (Supplementary formed lymphoblastoid cells to show an increased PARP Fig. S1A). Cleaved PARP was not due to degradation pro- cleavage compared with the FA-corrected counterparts or cesses as no difference in the band intensities (either upper with non-FA lymphoblastoid cells. This cleavage was par- or the lower) were shown in the lane with the phosphatase tially reduced by pretreating the cells with N-acetylcystein inhibitors (Supplementary Fig. S1B). In order confirm that (NAC; Fig. 1E). FA cells of I and M subgroups revealed sim- no IFN response occurs after FANCD2 siRNA transfection, ilar cleavage of PARP protein (data not shown). Immunoflu- we analyzed siRNA-transfected primary cells or cells stimu- orescence staining of the siRNA FANCD2–depleted cells lated with Poly-I-poly-C (positive control) by Western blot- (Supplementary Fig. S3); PD20, K561R, and PD20cor ting with eIF2α/P-eIF2α antibodies, or by performing (Supplementary Fig. S4); or FANCG, FANCL, and FAN- reverse transcription-PCR experiments to test the expression CA cells (supplementary Fig. S5) against cleaved anti-PARP of the IFN pathway–related gene STAT1 (Supplementary antibody revealed an increased intensity in PARP signaling. Fig. S2). To further expand our finding for other FA sub- Interestingly, pretreatment of cells with NAC decreased the

www.aacrjournals.org Mol Cancer Res; 8(1) January 2010 51

Lyakhovich and Surrallés

intensity of PARP signaling, suggesting that PARP cleavage cell. To correlate PARP cleavage with caspase3 activity, might be ROS mediated. Overall, these data show activation we tested the levels of active (cleaved) caspase3 in FA cells. and cleavage of PARP enzyme in FA cells without exogenous Western blot analyses revealed significant upregulation of damage. Although endogenous stress is probably a primary cleaved caspase-3 in the cell extracts of several FA comple- cause of caspase-3/PARP activation, the possibility of auto- mentation groups, but not in the corrected counterparts damage of FA cells due to exogenous stimuli cannot be com- (Fig. 2A). Cleaved caspase-8 and caspase-6 levels were also pletely disregarded. upregulated in some FA cell lines (data not shown). In- creased caspase-3 signaling was confirmed by immunoflu- Caspase-3 Is Upregulated in FA Cells orescence staining of PD20, PD20-corrected cells, K561R Caspase-3 plays a central role in the execution of the ap- cells (Fig. 2B), or primary fibroblasts transfected with optotic program and is primarily responsible for the cleav- siRNA or scRNA FANCD2 (Supplementary Fig. S6). age of PARP during cell death (28-30). At the same time, Moreover, we detected that the total level of PAR polymer caspase3 is a mediator of ROS accumulation within the synthesis, a proapoptotic marker, was also elevated in

FIGURE 3. Increased level of oxidized proteins and glutathione peroxidase activity in FA cells. A. Protein oxidation detection by the OxyBlotTM. B. Levels of oxidatively modified proteins quantified by scanning densitometry and expressed as fold increase versus normal controls using the TotaLab2.0 image program. Columns, mean (P < 0.02); bars, SD. C. Glutathione peroxidase activity in FA cells nontreated or pretreated with NAC was determined by NADPH oxidation at 340 nm. Columns, mean (n = 5 in each group; *, P < 0.01); bars, SD.

52 Mol Cancer Res; 8(1) January 2010 Molecular Cancer Research

Caspase-3/PARP Activation in Fanconi

FANCD2-deficient cells (Supplementary Fig. S7) or on oxidative stress, which may enhance the protein oxidative FANCD2 depletion (data not shown). To test caspase-3 profile. enzymatic activity in various FA complementation groups, we performed a colorimetric activity assay. In all FA cells, TNF-α Overproduction and MAPKs Upregulation in compared with the corrected counterparts, caspase-3 activ- FA Cells ity was higher (Fig. 2C). Moreover, the difference in the Recently, overproduction of TNF-α in FA has been re- FA-D2 group or in the primary cells depleted from ported both in vivo and in vitro (33-35). However, previous FANCD2 by siRNA were much higher than that of the experimental approaches did not establish whether TNF-α FA-upstream groups. Interestingly, only cells from the overproduction was directly linked to accumulation of FA-D2 group showed significant reduction of caspases-3 ROS in FA cells. To address this point, we compared the activity on treatment with NAC. This caspase assay was TNF-α level in the supernatants of exponentially growing alternatively validated by a chemiluminescence approach FA cells treated or untreated with ROS inhibitor. We ob- in which relative units of active caspase-3, measured after served that the TNF-α level was increased in all tested 30 minutes of incubation with the corresponding sub- supernatants of FA cells and that NAC significantly inhib- strate, were comparable with of colorimetric assay data (da- ited secretion of TNF-α (Fig. 4A). This observation sug- ta not shown). Thus, our results provide the first evidence gests that TNF-α secretion can be an early event of FA that caspase-3 activity is upregulated in FA cells without etiology and that the accumulation of this factor may even- exogenous DNA damage. tually lead to ROS accumulation and alter caspase-3/PARP Although caspase-3 and PARP cleavage are known mar- levels. At the same time, MAPKs may potentially activate kers of apoptosis, no confirmed data showing apoptosis of PARP cleavage and caspase-3 activity. Recently, aberrant intact FA cells have been described thus far. To compare activation of MAPKs was implicated in FANC-C pathway levels of apoptosis for different FA complementation loss-of-function and was also observed in FANCA-depleted groups, we performed a terminal deoxynucleotidyl trans- HeLa cells (33, 36). At the same time, ROS have been ferase-mediated dUTP nick end labeling–based apoptosis shown to activate several MAPKs and affect a wide array assay (Fig. 2D). Our data show that FA cells had higher of cellular processes, including proliferation, differentia- apoptotic levels when compared with corrected counter- tion, and apoptosis (37). Because phosphorylation of parts in the absence of DNA-damaging treatments. These ERK1/2 and P38 is a common event during oxidative findings were in parallel to the levels of caspase-3 activity. stress, we tested the activation of these kinases in FA cells Because pretreatment with ROS scavenger decreased by Western blotting. We found that in all tested FA cells, the number of apoptotic cells (Fig. 2D), PARP cleavage the levels of p-P38 and p-ERK1/2 were higher than in the (Fig. 1E), and caspases-3 activity (Fig. 2C), we concluded corrected counterparts (Fig. 4B). Pretreatment with ROS that moderate oxidative stress is a common property of FA inhibitor significantly reduced MAPK phosphorylation. cells growing in the absence of damaging factors. Similar observations were obtained by the immunofluorescent staining of FANCD2-deficient and FANCD2-profi- Protein Oxidation and ROS Accumulation in FA Cells cient cells (Supplementary Fig. S7A). Importantly, ROS have been implicated in PCD, cancer develop- increased upregulation of both pERK1/2 and p-P38 on ment, and aging (31) but can also be produced in normal FANCD2 depletion, and partial reduction of ERK1/2 cells by mitochondrial electron transport, cellular redox phosphorylation after NAC treatment was observed in pri- system, or immune responses (32). Proteins oxidized by mary fibroblasts as shown by immunofluorescence staining ROS often result in the incorporation of carbonyl groups (Supplementary Fig. S8). Thus, we suggest ROS-depen- into corresponding side chains. Therefore, we assessed the dent upregulation of pERK1/2 in FA cells. It may be pos- efficacy of oxidization processes in FA cells by measuring sible that the loss of function in FA genes results in TNF-α the amount of ROS-modified proteins. Our data on Oxy- oversecretion, which, in turn, activates MAPK by a ROS- Blot analyses show that the FA cells undergo oxidation dependent feedback loop and then triggers caspases-3/ more easily than corresponding corrected counterparts PARP cascade. Therefore, blocking TNF-α production (Fig. 3A and B). The same is true for primary cells con- may affect the expression of active caspase-3. To test this verted to FANCD2-like phenotype by siRNA. Moreover, hypothesis, we pretreated FA cells with an ERK inhibitor because NAC-pretreated samples partially reduce the level PD98059, which was earlier proved to reduce TNF-α pro- of oxidized proteins, accumulation of ROS seems to play a duction in FA-C cells (33). We showed that inhibition of role in FA cells. We then tested the activity of cellular glu- pERK1/2, on exposure to PD98059, was accompanied by tathione peroxidase, a member of the GPx enzymes that the reduction of cleaved caspase-3 level in FA cells serves as a quencher to protect cells from ROS damage. (Fig. 4C). Because ERK1/2 inhibition suppresses TNF-α We showed that the GPx activity in FA cells is impaired production, it may also reduce ROS level and, consequent- comparatively to the corrected counterparts and that pre- ly, decrease caspase-3 activity. Thus, it is unclear if endog- treatment with NAC only slightly affects this activity enously accumulated ROS increase the production of (Fig. 3C). Similar effects were found by measuring catalase TNF-α followed by triggering TNF receptors, accu- activity in FA cells (data not shown). These facts suggest mulating additional ROS, and increasing apoptotic mar- that FA cells have impaired protective properties against kers, or whether oversecretion of TNF-α–mediated ROS

www.aacrjournals.org Mol Cancer Res; 8(1) January 2010 53

Lyakhovich and Surrallés

FIGURE 4. Disruption of NAC-dependent TNF-α overproduction and MAPK activity in FA cells leads to caspases-3 downregulation. A. Overproduction of TNF-α in the supernatant of exponentially growing FA cells with or without NAC. Columns, mean (P < 0.05); bars, SEM. B. Immunoblots showing different MAPK profiles in FA cells. C. Immunoblots showing inhibition of ERK1/2, pERK1/2, and cleaved caspase-3 by pretreatment of FA cells with ERK inhibitor PD98059. D. Immunoblots (top) showing cleaved caspase-3 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) levels of FA-D2−/− (PD20) or FA-D2−/−–corrected (PD20cor) with the digitized signals (arrange as pairs of % redistribution). Bottom, cells pretreated with 1 μmol/L TNFR inhibitor WP9QY in the presence or absence of NAC. Corresponding cell replicas were loaded with H2-DCFDA (5 μmol/L) for 30 min and the DCF fluorescence (intracellular ROS) was determined. Data are expressed as a percentage of the untreated control; P < 0.05.

accumulation precedes these events. To rule out this possi- Discussion bility, we tested whether caspase-3 and ROS accumulation are dependent on TNFR1 by blocking TNFR1 with a spe- Evidence of FA sensitivity to oxygen was first provided cific peptidomimetic WP9QY. FANCD2-deficient or cor- by Joenje et al. (38), who showed that the chromosomal rected cells treated or untreated with ROS inhibitor were instability of primary FA cells could be reduced if grown incubated with WP9QY, and the levels of cleaved caspase-3 at lowered oxygen tension (39). Data obtained from mouse and intracellular ROS have been measured (Fig. 4D). Our models provided strong genetic evidence that FANCC−/− results show that blocking TNF receptor does affect caspase-3 cells are hypersensitive to endogenously generated oxidants, activity in D2−/−,butnotinD2−/−–corrected cells, sug- although it was unknown whether the molecular mecha- gesting that caspase-3 response is TNFα/TNFR1 mediated. nism responsible for this hypersensitivity was due to altered Intracellular ROS level is low in both cell lines and is redox signaling (40, 41). In turn, our data confirm that TNFα/TNFR1 dependent for D2−/− but not for D2−/−– ROS are playing a role in FA cells and that caspase-3/PARP corrected cells. Therefore, at least for the D2−/− subgroup, enzymes are activated along with MAPKs without exposure it is unlikely that ROS accumulation plays a primary role in of FA cells to exogenous damaging factors. apoptotic events. Rather, it is the overproduction of TNFα Recent studies of different human cDNA libraries, that causes caspase-3/PARP response. using a yeast two-hybrid system, identified interactors of

54 Mol Cancer Res; 8(1) January 2010 Molecular Cancer Research

Caspase-3/PARP Activation in Fanconi

FANCA, FANCC, and FANCG that involve proteins of unclear why the increase of such apoptotic markers as cas- oxidative metabolism groups (42). Moreover, works of Pa- pases-3, caspase-8, PARP, and PAR does not initiate apo- gano et al. (13, 15) suggested that the effects of commonly ptosis in FA cells. It should be noticed, however, that most used damaging agents, e.g., mitomycin C or cis-platin, may the cells used in our study were SV40-transformed cell equally activate both DNA repair/damage and oxidative- lines. Therefore, using primary mice or human FA cells damage response in FA cells. Potential sources of endoge- should rule out the possibility of nonspecific effects due nous stresses may involve interstrand cross-links formed by to immortalization. It is also possible that FA genetic back- products of lipid peroxidation and other forms of oxidative ground possess some compensatory factors that prevent ei- DNA damage (43). ther ROS increase or caspase-3–driven apoptosis or both. Some other factors playing an important role in aging Therefore, the challenge for future studies is to find out the and anemia include inflammatory ROS that are linked relationship between these processes and the different FA with DNA damage–induced premature senescence in he- intermediates. On a practical side, deeper understanding of matopoietic stem and progenitor cells. This seems to be why FA cells encounter oxidative stress in the absence of mediated through TNF-α–induced senescence that corre- DNA-damaging agents provide a strong link to a possible lates with the accumulation of ROS and oxidative DNA therapy for FA. damage (44). Interestingly, FANCC seems to be involved both in the protection of cells against oxidative damage Disclosure of Potential Conflicts of Interest and in the control of TNF-α activity (45). Cytoplasmic FANCC interacts with heat shock protein 70 to protect No potential conflicts of interest were disclosed. cells from IFN-γ/TNF-α–induced apoptosis (46). Such activation may involve transforming growth factor β1, Acknowledgments which induces concomitant activation of a caspase signaling cascade, in response to oxidative damage, followed by We thank Dr. Robert Grodzick (Thomas Steitz' PARP cleavage and p53-independent apoptosis (47). All laboratory, Yale University) for the careful reading of the these facts may potentially bridge the activity of caspases above manuscript and improving its quality Dr.Oehler in FA cells throughout oxidative stress and, more impor- (Medical University of Vienna) for providing access to tantly, provide some platform for explaining proapoptotic Typhoon scanning facility. nature of FA in the absence of mutagenic stress. Observa- tions by us and others of MAPKs activation in FA cells, Grant Support along with the previous findings of TNF-α release, may provide an explanation for caspase-3 activity and PARP Fanconi anemia research in Surrallés' laboratory is cleavage. One possible scenario that explains this phenom- funded by the Generalitat de Catalunya (2009 SGR enon goes to the recent findings of Rosselli's group (33) 489), ICREA-Academia program, Genoma España Foun- where they showed that ERK inhibition reduced TNF-α dation (FANCOGENE project), La Caixa Foundation oversecretion in FA-C cells. This was in parallel with the Oncology Program (BM05-67-0), the Spanish Ministry decreased matrix metalloproteinase-7 promoter activity. In of Science and Innovation (projects FIS06-1099, CI- our study, we reduced the level of cleaved caspases-3 by BER-ER CB06/07/0023, SAF2006-3440, and SAF2009- inhibiting ERK in both FANCD2-deficient and 11936), the Commission of the European Union (project FANCD2-proficient cells. If this is true for other FA cells, FI6R-CT-2003-508842), and the European Regional De- then the disruption of ERK (and possibly other MAPKs), velopment Funds. Centre for Biomedical Research on Rare followed by the reduction of TNF-α, may regulate caspases Diseases is an initiative of the Instituto de Salud Carlos III, triggering. We also show that blocking TNFR is accompa- Spain. nied by a decrease of cleaved caspase-3 level in D2−/− cells, The costs of publication of this article were defrayed in but not in D2−/−–corrected cells, and that intracellular part by the payment of page charges. This article must ROS level is TNFα/TNFR1 dependent for D2−/− but therefore be hereby marked advertisement in accordance not for D2−/−–corrected cells. This suggests that caspase- with 18 U.S.C. Section 1734 solely to indicate this fact. 3responseisTNFα/TNFR1 mediated and overproduc- Received 8/14/09; revised 11/19/09; accepted 11/19/09; tion of TNFα precedes ROS accumulation. Yet it is still published OnlineFirst 1/12/10.

References 1. Dokal I, Vulliamy T. Inherited aplastic anaemias/bone marrow failure 4. Mathew CG. Fanconi anaemia genes and susceptibility to cancer. syndromes. Blood 2008;22:141–53. Oncogene 2006;25:5875–84. 2. Wang W. Emergence of a DNA-damage response network consist- 5. Huang TT, D'Andrea AD. Regulation of DNA repair by ubiquitylation. ing of Fanconi anaemia and BRCA proteins. Nat Rev Genet 2007;8: Nat Rev Mol Cell Biol 2006;8:339–47. 735–48. 6. Kennedy RD, D'Andrea AD. The Fanconi Anemia/BRCA pathway: 3. Patel KJ, Joenje H. Fanconi anemia and DNA replication repair. DNA new faces in the crowd. Genes Dev 2005;19:2925–40. Repair (Amst) 2007;6:885–90. 7. Fagerlie S, Lensch MW, Pang Q, Bagby GC. The Fanconi anemia

www.aacrjournals.org Mol Cancer Res; 8(1) January 2010 55

Lyakhovich and Surrallés

group C gene product: signaling functions in hematopoietic cells. 28. Alnemri E, Livingston D, Nicholson D, et al. Human ICE/CED-3 pro- Exp. Hematol 2001;29:1371–81. tease nomenclature. J Cell 1996;87:171. 8. Alter B, Young N. In: Nathan D, Oski F, editors. Hematology of Infan- 29. Nicholson DW, Ali A, Thornberry NA, et al. Identification and inhibi- cy and Childhood. New York: 1998. p. 27–335. tion of the ICE/CED-3 protease necessary for mammalian apoptosis. 9. Saunders WB. In: Young NS, editor. Bone Marrow Failure Syn- Nature 1996;376:37–43. dromes. Philadelphia: 2000. p. 47–68. 30. Tewari M, Quan LT, O'Rourke K, et al. Yama/CPP32 β, a mammalian 10. Cimino F, Esposito F, Ammendola R, Russo T. Gene regulation by homolog of CED-3, is a CrmA-inhibitable protease that cleaves the reactive oxygen species. Curr Top Cell Regul 1997;35:123–48. death substrate poly(ADP-ribose) polymerase. Cell 1995;81:801–9. 11. Dalton TP, Shertzer HG, Puga A. Regulation of gene expression by 31. Stadtman ER. Protein oxidation and aging. Science 1992;257:1220. reactive oxygen. Annu Rev Pharmacol Toxicol 1999;39:67–101. 32. Halliwell B. Free radicals, antioxidants, and human disease: curiosity, 12. Du W, Adam Z, Rani R, Zhang X, Pang Q. Oxidative stress in Fanconi cause, or consequence? Lancet 1994;344:721–4. anemia hematopoiesis and disease progression. Antioxid Redox Sig- 33. Briot D, Macé-Aimé G, Subra F, Rosselli F. Aberrant activation of nal 2008;10:1909–21. stress-response pathways leads to TNF-α oversecretion in Fanconi 13. Pagano G, Degan P, d'Ischia , et al. Oxidative stress as a multiple anemia. Blood 2008;111:1913–23. effector in Fanconi anaemia clinical phenotype. Eur J Haematol 34. Secchiero P, Zauli G. Tumor-necrosis-factor-related apoptosis-in- 2005;75:93–100. ducing ligand and the regulation of hematopoiesis. J Clin Invest 14. Cohen-Haguenauer O, Péault B, Bauche C, et al. In vivo repopulation 2007;117:3283–95. ability of genetically corrected bone marrow cells from Fanconi ane- 35. Sejas DP, Rani R, Qiu Y, et al. Inflammatory reactive oxygen species- mia patients. Proc Natl Acad Sci U S A 2006;103:2340–45. mediated hemopoietic suppression in Fancc-deficient mice. J Immu- 15. Pagano G, Degan P, d'Ischia M, et al. Gender- and age-related dis- nol 2007;178:5277–87. tinctions for the in vivo prooxidant state in Fanconi anaemia patients. 36. Ferrer M, Izeboud T, Ferreira GC, et al. Cisplatin triggers apoptotic or Carcinogenesis 2004;25:1899–9. nonapoptotic cell death in Fanconi anemia lymphoblasts in a con- 16. Park SJ, Ciccone SL, Beck BD, et al. Oxidative stress/damage in- centration-dependent manner. Exp Cell Res 2003;286:381–95. duces multimerization and interaction of Fanconi anemia proteins. 37. Chang L, Karin M. Mammalian MAP kinase signalling cascades. Na- J Biol Chem 2004;279:30053–59. ture 2001;410:37–40. 17. Pearl-Yafe M, Halperin D, Scheuerman O, Fabian I. The p38 pathway 38. Joenje H, Arwert F, Eriksson A, de Koning H, Oostra A. Oxygen-de- partially mediates caspase-3 activation induced by reactive oxygen pendence of chromosomal aberrations in Fanconi's anaemia. Nature species in Fanconi anemia C cells. Biochem Pharmacol 2004;67: 1981;290:142–43. 539–46. 39. Guillouf C, Wang TS, Liu J, Walsh CE, Poirier GG, Moustacchi E, 18. Rathbun RK, Christianson TA, Faulkner GR, et al. Interferon-γ-in- Rosselli F. Fanconi anemia C protein acts at a switch between apo- duced apoptotic responses of Fanconi anemia group C hematopoi- ptosis and necrosis in mitomycin C-induced cell death. Exp Cell Res etic progenitor cells involve caspase 8-dependent activation of 1999;246:384–94. caspase 3 family members. Blood 2000;96:4204–11. 40. Joenje H, Youssoufian H, Kruyt F, dos Santos C, Wevrick R, Buchwald 19. Salvesen GC, Riedl SJ. Caspase mechanisms. Adv Exp Med Biol M. Expression of the Fanconi anemia gene FAC in human cell lines: 2008;615:13–23. lack of effect of oxygen tension. Blood Cells Mol Dis 1995;21:182–91. 20. Schreiber V, Dantzer F, Ame JC, de Murcia G. Poly(ADP-ribose): 41. Hadjur S, Ung K, Wadsworth I, et al. Defective hematopoiesis and novel functions for an old molecule. Nat Rev Mol Cell Biol 2006; hepatic steatosis in mice with combined deficiencies of the genes 7:517–28. encoding Fancc and Cu/Zn superoxide dismutase. Blood 1998;98: 21. Decker P, Muller S. Modulating poly (ADP-ribose) polymerase activ- 1003–11. ity: potential for the prevention and therapy of pathogenic situations 42. Reuter TY, Medhurst AL, Waisfisz Q, et al. Yeast two-hybrid screens involving DNA damage and oxidative stress. Curr Pharm Biotechnol imply involvement of Fanconi anemia proteins in transcription regu- 2002;3:275–83. lation, cell signaling, oxidative metabolism, and cellular transport. 22. Bogliolo M, Lyakhovich A, Callen E, et al. Histone H2AX and Fanconi Exp Cell Res 2003;289:211–21. anemia FANCD2 function in the same pathway to maintain chromo- 43. Pang Q, Andreassen PR. Fanconi anemia proteins and endogenous some stability. EMBO J 2007;26:1340–51. stresses. Mutat Res 2009;668:42–53. 23. Flohe L, Gunzler WA. Assays of glutathione peroxidase. Methods En- 44. Zhang X, Sejas DP, Qiu Y, Williams DA, Pang Q. Inflammatory ROS zymol 1984;105:114–21. promote and cooperate with the Fanconi anemia mutation for hema- 24. Aebi H. Catalase in vitro. Methods Enzymol 1984;105:121–26. topoietic senescence. J Cell Sci 2002;120:1572–83. 25. Ramirez MH, Adelfalk C, Kontou M, Hirsch-Kauffmann M, Schweiger 45. Pang Q, Christianson TA, Keeble W, et al. The Fanconi anemia com- M. The cellular control enzyme polyADP ribosyl transferase is elimi- plementation group C gene product: structural evidence of multi- nated in cultured Fanconi anemia fibroblasts at confluency. Biol functionality. Blood 2001;98:1392–1. Chem 2003;384:169–74. 46. Pang Q, Keeble W, Christianson TA, Faulkner GR, Bagby GC. 26. Lyakhovich A, Surrallés J. New roads to FA/BRCA pathway: H2AX. FANCC interacts with Hsp70 to protect hematopoietic cells from Cell Cycle 2007;6:1019–23. IFN-γ/TNF-α-mediated cytotoxicity. EMBO J 2001;20:4478–89. 27. Aoufouchi S, Shall A. Regulation by phosphorylation of Xenopus lae- 47. Albright CD, Borgman C, Craciunescu CN. Activation of a caspase- vis poly(ADP-ribose) polymerase enzyme activity during oocyte mat- dependent oxidative damage response mediates TGFβ1 apoptosis uration. Biochem J 1997;325:543–51. in rat hepatocytes. Exp Mol Pathol 2003;74:256–61.

56 Mol Cancer Res; 8(1) January 2010 Molecular Cancer Research

Constitutive Activation of Caspase-3 and Poly ADP Ribose Polymerase Cleavage in Fanconi Anemia Cells

Alex Lyakhovich and Jordi Surrallés

Mol Cancer Res 2010;8:46-56. Published OnlineFirst January 12, 2010.

Updated version Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-09-0373

Supplementary Access the most recent supplemental material at: Material http://mcr.aacrjournals.org/content/suppl/2010/01/08/8.1.46.DC1

Cited articles This article cites 45 articles, 10 of which you can access for free at: http://mcr.aacrjournals.org/content/8/1/46.full#ref-list-1

Citing articles This article has been cited by 2 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/8/1/46.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/8/1/46. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.