FEBS 29179 FEBS Letters 579 (2005) 615–620

Differentiative pathway activated by 3-aminobenzamide, an inhibitorof PARP, in human osteosarcoma MG-63 cells

A. De Blasioa, C. Messinab, A. Santullib, V. Manganoa, E. Di Leonardoa, A. D�Anneoa,G. Tesorierea,1, R. Ventoa,*,1

a Dipartimento di Biologia Cellulare e dello Sviluppo, Sezione di Biochimica, Universita di Palermo, Policlinico, via del Vespro 129, 90127 Palermo, Italyb Istituto di Biologia Marina, Consorzio Universitario Provincia di Trapani, Lungomare Dante Alighieri, 91016 Trapani, Italy

Received 28 October 2004; revised 4 December 2004; accepted 4 December 2004

Available online 24 December 2004

Edited by Robert Barouki

Abstract This study describes the molecular mechanism bywhich treatment with 3-AB, a potent inhibitor of PARP, allowshuman osteosarcoma MG-63 cells to restrict growth and enterdifferentiation. Our findings show that in MG-63 cells, aberrantgene expression keeps Rb protein constitutively inactivatedthrough hyperphosphorylation and this promotes uncontrolledproliferation of the cells. After 3-AB-treatment, the poly(ADP-ribosyl)ation of nuclear proteins markedly decreases and this re-sults in an increase in both the hypophosphorylated active formof Rb and pRb/E2F complexes. These effects are accompaniedby G1 arrest, downregulation of gene products required for pro-liferation (cyclin D1, b-catenin, c-Jun, c-Myc and Id2) andupregulation of those implicated in the osteoblastic differentia-tion (p21/Waf1, osteopontin, osteocalcin, type I collagen, N-cadherins and alkaline phosphatase). Our study suggests thatuse of PARP inhibitors may induce a remodeling of chromatinwith the reprogramming of gene expression and the activationof differentiation.� 2004 Federation of European Biochemical Societies. Publishedby Elsevier B.V. All rights reserved.

Keywords: Osteosarcoma; Differentiation; PARP; Rb protein;Cell cycle; 3-AB

1. Introduction

In eukaryotic cells, PARP catalyzes poly(ADP-ribosyl)ation

reactions, which transfer the poly(ADP-ribose) moiety to his-

tone and non-histone proteins [1]. It has been suggested that

PARP can be critical for gene regulation in vivo by creating

an anionic poly(ADPribose) matrix that binds histones, there-

by promoting the decondensation of higher-order chromatin

structures [2].

Accumulated data suggest that PARP takes part mainly in

nuclear events, including DNA repair, cell cycle [3], apoptosis

[4], and maintenance of chromosomal stability [5]. Activation

Abbreviations: PARP, poly(ADP-ribose)polymerase; 3-AB, 3-aminob-enzamide

*Corresponding author. Fax: +39 091 6552449.E-mail address: rvento@unipa.it (R. Vento).

1 These authors contributed equally to the conception and design ofthe study.

0014-5793/$30.00 � 2004 Federation of European Biochemical Societies. Pu

doi:10.1016/j.febslet.2004.12.028

of PARP has been implicated in a number of pathologies such

as stroke, myocardial ischemia, diabetes and central nervous

system pathologies. Therefore, inhibition of PARP by pharma-

cological agents has been suggested as a useful treatment for

these diseases [6,7].

We have recently demonstrated [8] that in human osteosar-

coma cells, treatment with 3-aminobenzamide (3-AB), a potent

inhibitor of PARP, activates a conflicting signal which deter-

mines a different final destiny for cells that express functional

Rb protein from those that do not. In Saos-2 cells lacking

Rb function, 3-AB treatment drives cells through an apoptotic

pathway, while in MG-63 cells expressing wild-type Rb, 3-AB

treatment drives cells to differentiate into osteocyte-like cells.

Our aim in this study was to elucidate the molecular mecha-

nism by which PARP inhibition allows human osteosarcoma

MG-63 cells to restrict growth and enter differentiation. We

demonstrate for the first time that inhibiting PARP in human

osteosarcoma MG-63 cells results in the activation of func-

tional Rb with downregulation of gene products required for

proliferation and upregulation of those required for differenti-

ation. These findings suggest that the reduction in ADP-

ribosylation of nuclear proteins induced by 3-AB treatment

may determine a modulation in chromatin structure which in

turn may result in a reprogramming of gene expression.

2. Materials and methods

2.1. Cell culture and reagentsHuman osteosarcoma MG-63 cells were routinely grown as previ-

ously described [8]. The cells were treated with either vehicle or 10mM 3-AB for the times indicated in the Figures. Material for RT-PCR was from Applied Biosystems (CA, USA). Anti-poly(ADP-ri-bose) monoclonal antibody 10H was kindly supplied by Prof. Alexan-der Burkle, University of New Castle upon Tyne UK; Phosphoplus Rb(Ser-780, Ser-795, Ser-807/811) antibodies were from Cell SignalingTechnology; all other antibodies were from Santa Cruz Biotechnology,Inc. Other materials and reagents were from Sigma–Aldrich.

2.2. Flow cytometry analysisMG-63 cells were incubated in the absence or presence of 3-AB, har-

vested, centrifuged, resuspended in 400 ll of PBS containing 10 lg/mlRNase, and stained with 50 lg/ml propidium iodide. Samples of 2–4 · 105 cells were then analyzed for DNA histograms and cell cyclephase distributions by flow cytometer using a Coulter Epics XL (Beck-man Coulter). To evaluate whether 3-AB-induced G-1 phase arrest wasreversible, after 48 h of 3-AB treatment the medium was removed, andcells were washed and refed with fresh control medium without 3-AB.After an additional 24, 48, or 72 h following the removal of 3-AB, cellswere analyzed for DNA histograms and cell cycle phase distributions.

blished by Elsevier B.V. All rights reserved.

616 A. De Blasio et al. / FEBS Letters 579 (2005) 615–620

2.3. Nuclear extract10 · 106 cells were incubated in the absence or presence of 3-AB,

harvested and the nuclei were extracted as reported by Rabinovichet al. [9].

2.4. Western blot analysis and immunoprecipitationWhole cell lysates were obtained by disrupting cells in lysis buffer [8].

Equal amounts of proteins (60 lg) were separated by SDS–PAGE andtransferred onto nitrocellulose membranes. In each experiment, Pon-ceau S staining was performed to check that closely comparableamounts of proteins had been loaded and transferred in each lane.The filters were then probed with the specific antibodies. Reprobingwas performed following incubation of membranes in stripping buffer(100 mM b-mercaptoethanol, 2% SDS and 62.5 mM Tris–Cl, pH 6.7)overnight. Bands were quantified by densitometric analysis [8]. Forimmunoprecipitations, 400 lg of nuclear extracts was incubated with0.4 lg of the appropriate primary antibody for 3 h and then incubatedovernight with 30 ll of protein A/G plus agarose beads. The immuno-complexes were washed with lysis buffer, boiled in SDS sample bufferand submitted to Western blotting.

2.5. RT-PCR analysisRNA was isolated from MG-63 cells using RNeasy mini kit (Qia-

gen). cDNA was amplified from 1 lg of RNA and PCR was performedas previously reported [8]. The amplified products were resolved byagarose gel electrophoresis (1% agarose–0.5 lg/ml ethidium bromide)and then photographed and scanned into Adobe Photoshop. GAPDHwas employed as an internal control. The following primer pairs(Proligo USA) were used. b-catenin: 285 bp, 5 0-CGTGGACAATG-GCTACTCAAGC-30 sense/ 5 0-TCTGAGCTCGAGTCATTGCA-TAC-30 antisense; cyclin D1: 587 bp, 50-TGGATGCTGGAGGTCTG-CGAGGAA-3 0 sense/5 0-GCTTGACTCCAGCAGGGCTTCGAT-3 0

antisense; GAPDH: 200 bp, 5 0-TGACATCAAGAAGGTGA-3sense/�5 0-TCCACCACCCTGTTGCTGTA-3 0 antisense; ID2: 207 bp,5 0-CATCTTGGACCTGCAGATCG-3 0 sense/5 0-ATGAACACCGCTTATTCAGC-3 0 antisense; c-Jun: 316 bp, 5 0-GGAAACGACCTTC-TATGACG-3 0 sense/5 0-GAACCCCTCCTGCTCATCTGTCACGTTCTT-3 0 antisense; c-Myc: 239 bp, 5 0-GACCACAAG GCCCTCAG-TAC-3 0 sense/5 0-GTGGATGGGAAGGCATCGTT-3 0 antisense;osteocalcin: 241 bp, 5 0-CTCACACTCCTCGCCCTATT-3 0 sense/ 5 0-GGTCAGCCAACTCGTCACAG-30 antisense; osteopontin: 348 bp,5 0-CCAAGTAAGTCCAACGAAAG-3 0 sense/5 0-GGTGATGTCCTCGTCTGTA-3 0 antisense; type I collagen: 219 bp, 5 0-ATGTCTAGGGTCTAGACATGTTCA-30 sense/5 0-CCTTGCCGTTGTCGCAGACG-3 0 antisense.

2.6. Assay of alkaline phosphatase activityALP activity was evaluated using the Sigma Diagnostics Alkaline

Phosphatase kits (Sigma-Aldrich) intended for the histochemical semi-quantitative demonstration (procedure No. 85), according to the man-ufacturer�s instructions.

2.7. Reversibility of MG-63 effectsMG-63 cells were treated with 3-AB and after 48 h the medium was

removed and cells were washed and refed with fresh control mediumwithout 3-AB. After an additional 24, 48 or 72 h following the removalof 3-AB, cytometry, Western blotting and RT/PCR analysis wereperformed.

Fig. 1. Pattern of ADP-ribosylation in either whole cells lysates ornuclear extracts of MG-63 cells. Effects of 3-AB. The cells wereincubated in the absence or presence of 3-AB, after which whole celllysates (60 lg) or nuclear extracts (30 lg) were resolved by 10% (A) or12% (B, C) SDS–PAGE and immunoblotted with antibodies to eitherpoly-ADP-ribose (A and C), or H2b histone (B). Data are represen-tatives of four separate experiments.

3. Results

3.1. Pattern of ADP-ribosylation in MG-63 cell extracts.

Inhibitory effects of 3-AB

Our initial goal in this study was to ascertain whether the

treatment with 3-AB of human osteosarcoma MG-63 cells re-

sults in a decrease in the content of ADP ribosylated proteins.

The analysis was performed by Western blot of whole cell ex-

tracts prepared after treatment with 3-AB. Probing the filter

with anti-PAR, a monoclonal antibody which recognizes

poly(ADP-ribose)-modified proteins and which does not

cross-react with RNA, DNA or ADP-ribose monomers, a vast

array of poly(ADP-ribosyl)ated proteins appeared which sig-

nificantly decreased with the time of the 3-AB treatment

(Fig. 1A). When nuclear extracts were probed with either

anti-H2B histone antibody (Fig. 1B) or anti-PAR antibody

(Fig. 1C), the results suggested that histones may be among

the poly(ADP-ribosyl)ated proteins which appeared to signifi-

cantly decrease with 3-AB treatment.

3.2. 3-AB treatment induces morphological differentiation and

G1 phase arrest

The microscopy observation of MG-63 cells demonstrates

the changes induced, as compared with the control (Fig. 2A),

after 48 h of treatment with 3-AB (Fig. 2B). The potent reduc-

tion in cell number, accompanied by dramatic morphological

changes, is suggestive of differentiative attempts. Treated cells

appeared much larger than control, with a stellate morphol-

ogy, resembling osteocyte-bearing long cytoplasmatic protru-

sions which interconnect each other. Cytometry analysis

(Fig. 2C) confirms [8] that 3-AB treatment induces potent

changes in the cell cycle phases population with G1 arrest.

The effect of 3-AB was time-dependent with the percentage

of cells in G1 phase increasing by 8%, 18.4% and 24.4% after

16, 24 and 48 h of treatment, respectively. The analysis also

shows that the progressive G1 accumulation observed after

3-AB treatment was mostly associated with a concomitant de-

crease of cells in the S and G2-M phases of the cell cycle.

3.3. Rb inMG-63 cells is inactivated by hyperphosphorylation;

3-AB treatment induces the production of the

hypophosphorylated/active status of Rb and increases the

levels of the pRb/E2F complex

It is known that the G1 arrest of the cells depends on the

functional hypophosphorylated/active Rb form (pRb); pro-

gressive Rb phosphorylation leads to hyperphosphorylated/

inactive Rb form (ppRb) and simultaneous G1-S transition.

Hypophosphorylated Rb binds and sequesters cellular pro-

teins, most notably transcription factors of the E2F/DP family,

thereby regulating the transcription of genes [10]. We thus ex-

plored the functional status of Rb and the formation of pRb/

E2F complexes. Western blot analysis of whole cell lysates

probing the filter with antibody to the Rb (A/B pocket) (Fig.

Fig. 2. Microscopy and cytometry analysis of MG-63 cells treatedwith 3-AB. The cells were treated for 48 h with vehicle (A) or with 3-AB (B) and observed by phase contrast microscopy. Data arerepresentatives of seven separate experiments done in triplicate.Magnification ·400. In panel C, the cells were analyzed by DNA flowcytometry. Data are representatives of four separate experiments.

Fig. 3. Effects of 3-AB on the phosphorylated status of Rb and on thelevels of pRb/E2F complex. The cells were treated with 3-AB andwhole cell lysates (60 lg) were subjected to 10% SDS–PAGE andimmunoblotting (A). Filters were probed with antibodies to Rb (a), orphospho-Rb Ser-780 (b), Ser-795 (c), Ser-807/811 (d). The filter d wasstripped and reprobed with anti Rb antibody (e). In panel B, whole celllysates (60 lg) were resolved by 10% SDS–PAGE and immunoblottedwith anti-E2F1 (a) or anti-Rb (b) antibodies. In addition, nuclearextracts (400 lg protein each) were immunoprecipitated with anti Rbantibodies, resolved by 10% SDS–PAGE and blotted. The filter wasprobed with anti-E2F1 (c), stripped and reprobed with anti Rbantibody (d). Data are representatives of four separate experiments.

A. De Blasio et al. / FEBS Letters 579 (2005) 615–620 617

3A, a) resulted in multiple closely spaced bands of between

approximately 110–116 kDa. The bands might represent differ-

ent Rb phosphorylation forms, with the lower mobility bands

being more highly phosphorylated than the higher mobility

bands. 3-AB treatment induced a marked reduction in the lower

mobility bands with a simultaneous increase in the higher

mobility bands. Probing new filters with antibodies against

phospho-Rb serine-780 (Fig. 3A, b), phospho-Rb serine-795

(Fig. 3A, c) and phospho-Rb serine-807/811 (Fig. 3A, d) dem-

onstrated that Rb protein in MG-63 cells is constitutively

hyperphosphorylated and that 3-AB treatment strongly de-

creases the amount of ppRb. Within 16 h of the addition of

3-AB to the cell culture, there was little decrease in ppRb

(not shown). The effect of 3-AB was progressively more evi-

dent after 24 h and 48 h of treatment and was evident on all

the phosphoserine residues studied, in particular on phospho-

serine 795. When the filter probed with anti-phospho-Rb ser-

ine-795 antibody was stripped and reprobed with the anti-Rb

antibody (Fig. 3A, e), it was evident that the lowering in the

ppRb form was accompanied by a concomitant increase in

the pRb form.

In light of these results, we then sought to determine whether

3-AB treatment induced changes in the pRb/E2F complex. Ini-

tially, nuclear extracts were submitted to Western blot analysis

and the filters were reacted with either anti E2F-1 (Fig. 3B, a)

or anti-Rb (Fig. 3B, b) antibodies. The results show that 3-AB

treatment did not change the level of E2F-1, while it markedly

modified the level of ppRb in favor of pRb. Thereafter, the nu-

clear extracts were immunoprecipitated with anti-Rb antibody,

resolved by western blot and probed with the anti E2F-1 anti-

body (Fig. 3B, c). The results demonstrated that 3-AB treat-

ment induced a significant increase in the E2F-1 component.

When the same filter was stripped and reprobed with the anti

Rb antibody (Fig. 3B, d), the results showed that, after 3-AB

treatment, most of the immunoprecipitated Rb is present in

the hypophosphorylated form. The findings suggest that 3-

AB treatment induces a significant increase in the pRb/E2F

complex.

3.4. 3-AB treatment decreases the expression of genes involved in

cell cycle progression and increases that of genes responsible

for osteoblast differentiation program

pRb phosphorylation occurs by cyclin-dependent kinases

(CDKs), with cyclin D1 playing crucial roles in favoring the

progression from G1 into the S phase which, in turn, is im-

peded by p21/Waf1, a potent inhibitor of CDKs [11].

We thus ascertained whether treatment with 3-AB induces

changes in the levels of these cell cycle regulators byWestern blot

(Fig. 4A) and RT-PCR (Fig. 4B) analysis of whole cell lysates.

The Figure demonstrates that the level of cyclin D1 was quite

high, while p21/Waf-1 was barely measurable. Treatment with

3-AB induced a considerable time-dependent decrease in the

Fig. 4. Effects of 3-AB on the expression of both various typical controllers of cell cycle and various markers of osteoblast differentiation. (A), equalamounts of whole cell lysates (60 lg) were subjected to SDS–PAGE and immunoblotting. Filters were probed with antibodies to cyclin D1, p21/Waf1, b-catenin, c-jun, ID2, c-Myc, osteopontin (OSP), N-cadherin (N-CAD), osteocalcin (OSC), and type I collagen. The data shown under eachblot are from the densitometry analysis. (B), RNA was extracted and analyzed by RT-PCR with primers for alkaline phosphatase (ALP), cyclin D1,c-Jun, b-catenin, ID2, c-Myc, type I collagen (coll), OPN, OSC or GAPDH (internal control). (C), hystochemical evaluation of ALP. Data arerepresentatives of five separate experiments.

618 A. De Blasio et al. / FEBS Letters 579 (2005) 615–620

level of cyclin D1 with a simultaneous increase in p21/Waf-1.

The results suggest that 3-AB induces molecular changes that

can inhibitG1/S transition. Since it is known that cyclinD1 tran-

scription can be regulated byb-catenin, we examined its level. As

Fig. 4 shows,MG-63 cells express high levels of b-catenin whichpotently decrease by 3-AB-treatment. Also, c-Jun (a gene in-

volved in cell immortalization in a number of cancer cells) can

be regulated by b-catenin and the figure shows significant levels

of c-Jun that markedly lowered after 3-AB-treatment.

To explore whether additional mechanisms can contribute to

the functional inactivation of pRb, we examined the expression

of the helix-loop-helix ID2 proteins, known to be involved in

the disturbed regulation of cell cycle by interfering with Rb

protein [12]. Western blot and RT-PCR analysis (Fig. 4) dem-

onstrated very high levels of ID2, which markedly decreased

after 3-AB treatment. Since ID2 operates, at least in part, un-

der the control of c-Myc [13], which, by raising ID2 levels, cir-

cumvent the block on cell cycle progression imposed by the Rb

pathway, we also checked for c-Myc involvement. The results

demonstrate that MG-63 cells even express significant levels of

c-Myc, which were downregulated after 3-AB-treatment.

We have previously shown [8] that in MG-63 cells, 3-AB

treatment induces morphological and biochemical changes

suggestive of osteocyte differentiation. Of the biochemical pat-

tern previously analyzed [8], we have shown that CD44 (the

osteopontin receptor, sensitive marker of osteocyte differentia-

tion) potently increases after 3-AB-treatment. Fig. 4 also dem-

onstrates that 3-AB treatment potently increases the

expression of ECM proteins and that of cell–cell adhesion pro-

teins which are implicated in the expression of the differentia-

tion bone-program. Western blot (A) and RT-PCR (B)

analysis show that 48 h of 3-AB-treatment significantly in-

creases the expression of both osteopontin (5.1 fold), one of

the most prominent extracellular matrix proteins synthesized

by osteoblastic cells during bone development [14], and type

I collagen (3.2 fold), another major organic component of

bone ECM, which is fundamental to the development expres-

sion of the bone cell phenotype. Analysis of osteocalcin expres-

sion showed a 3.8-fold-increase that further confirmed the

osteogenic phenotype induced by 3-AB treatment. Cell–cell

adhesion mediated by N-cadherins plays an essential role in

the control of human osteoblast differentiation and bone for-

mation [15], and Fig. 4A shows that after 3-AB treatment N-

cadherins increased by 2.2-fold. Interestingly, RT/PCR (Fig.

4B) and histochemical semiquantitative dosage (Fig. 4C) show

that prolonged times of 3-AB treatment (216 h) even stimu-

lated the expression of alkaline phosphatase, the most typical

marker of osteocyte differentiation.

Fig. 5. Reversibility of 3-AB effects. Cytometry analysis of cell cycle(A) and expression of cyclin D1 (B), after refeding in the absence of 3-AB. Data are representatives of three separate experiments.

A. De Blasio et al. / FEBS Letters 579 (2005) 615–620 619

3.5. Reversibility of the effects induced by 3-AB: cell cycle and

cyclin D1analysis

In some experiments MG-63 cells were treated with 3-AB

and after 48 h, the medium was removed and cells were washed

and refed with fresh control medium without 3-AB. After an

additional 24, 48, or 72 h following the removal of 3-AB, cells

were analyzed either by cytometry for cell cycle (Fig. 5A), or

(after protein and RNA extraction) by Western blot and RT/

PCR for cyclin D1 (Fig. 5B). The results show that the effect

induced by 48 h of 3-AB-treatment seemed not to be irrevers-

ible as it was entirely reversed after 78 h of refeding.

4. Discussion

This paper describes the molecular mechanism by which

treatment with 3-AB, an inhibitor of PARP, allows human

osteosarcoma MG-63 cells to restrict growth and enter differ-

entiation. We provide evidence that Rb protein, the main

brake on progress around the cell-division cycle, is constitu-

tively inactivated by hyperphosphorylation. As a consequence,

cells progress uncontrolled through the cell-cycle. The findings

show that 3-AB treatment potently reduced ADP-ribosylation

of nuclear proteins, among which the histones. The reduction

in ADP-ribosylation was accompanied by an accumulation of

the hypophosphorylated Rb form, with a simultaneous potent

downregulation of the gene products responsible for cell prolif-

eration (cyclin D1, b-catenin and c-Jun). This activated a cell-

cycle blockade mechanism which involved the p21/E2F system.

This paper also shows that MG-63 cells express very high lev-

els of Id2 proteins. These proteins are known to potently inhi-

bit differentiation in multiple cell types and to corroborate the

cell cycle machinery either by binding to the hypophosphory-

lated Rb protein and abolishing its growth-suppressing activity

[16] or by antagonizing the p21/Waf1 encoding gene [17].

Treatment with 3-AB potently reduced ID2 expression and

this was accompanied by a potent increase in p21/WAF1 lev-

els. The treatment also induced downregulation of N-Myc,

known to upregulate ID2 gene [13]. Finally, treatment with

3-AB induced a significant increase in the expression of osteo-

pontin, osteocalcin, type-1 collagen, N-cadherin and ALP, the

production of which is considered to be fundamental to the

developmental expression of the bone cell phenotype.

In conclusion, this paper suggests a scenario in which the

possible excessive ADP-ribosylation of nuclear proteins may

cause the aberrant gene expression that keeps Rb inactive

through hyperphosphorylation and MG-63 cells continuously

cycling. The cancellation of ADP-ribosylation of nuclear pro-

teins determined by 3-AB treatment may induce modulation in

chromatin structure, which in turn may result in gene expres-

sion reprogramming and the induction of differentiation.

Localized changes in chromatin structure are a key event in

the transcriptional regulation of genes. Histone tails are sub-

ject to multiple post-translational modifications such as acety-

lation, phosphorylation, ubiquitination, methylation, and

poly-ADP-ribosylation, which play a role in transcriptional

regulation [18]. It has been shown that PARP can modify nu-

clear proteins creating an anionic poly(ADP-ribose) matrix,

which promotes the remodeling of chromatin [2]. Further-

more, excessive PARP activity may be implicated in the path-

ogenesis of numerous clinical conditions including cancer [19].

PARP inhibitors have recently been shown to be effective in

sensitizing cancer cells to drug-induced apoptosis, and treat-

ment of cancer with PARP inhibitors in combination with

other agents thus appears to be a promising new therapeutic

approach [20]. Osteosarcoma is a common malignant tumor

for which the current drug treatment is chemotherapy. This

commonly involves a combination of multiple anticancer

drugs, such as doxorubicin, cisplatin, and/or methotrexate.

However, more than 30% of patients are resistant to these che-

motherapy regimens [21]. To our knowledge, our paper is the

first demonstration that treatment with 3-AB alone can induce

antiproliferative and differentiative effects in cancer cells. We

propose that PARP inhibition may be a novel approach for

the treatment of osteosarcoma.

Acknowledgments: This study was supported by grants from ‘‘Minis-tero dell�Istruzione, dell�Universita e della Ricerca’’. We are gratefulto Professor Burkle for providing the antibodies to poly-ADP-ribose.We thank Prof. G. Calvaruso for his help in cytofluorimetry analysis.

References

[1] Soldatenkov, V.A. and Potaman, V.N. (2004) DNA-bindingproperties of poly(ADP-ribose) polymerase: a target for antican-cer therapy. Curr. Drug Targets. 5, 357–365.

[2] Kraus, W.L. and Lis, J.T. (2003) PARP goes transcription. Cell113, 677–683.

[3] Yung, T.M., Sato, S. and Satoh, M.S. (2004) Poly(ADP-ribosyl)ation as a DNA damage-induced post-translational mod-ification regulating poly(ADP-ribose) polymerase-1-topoisomer-ase I interaction. J. Biol. Chem. 279, 39686–39696.

[4] Ivana Scovassi, A. and Diederich, M. (2004) Modulation ofpoly(ADP-ribosylation) in apoptotic cells. Biochem. Pharmacol.68, 1041–1047.

[5] Kanai, M., Uchida, M., Hanai, S., Uematsu, N., Uchida, K. andMiwa, M. (2000) Poly(ADP-ribose) polymerase localizes to thecentrosomes and chromosomes. Biochem. Biophys. Res. Com-mun. 278, 385–389.

620 A. De Blasio et al. / FEBS Letters 579 (2005) 615–620

[6] Beneke, S., Diefenbach, J. and Burkle, A. (2004) Poly(ADP-ribosyl)ation inhibitors: promising drug candidates for a widevariety of pathophysiologic conditions. Int. J. Cancer 111, 813–818.

[7] Soos, J., Engelhardt, J.I., Siklos, L., Havas, L. and Majtenyi,K. (2004) The expression of PARP, NF-kappa B and parval-bumin is increased in Parkinson disease. Neuroreport 5, 1715–1718.

[8] De Blasio, A., Musmeci, M.T., Giuliano, M., Lauricella, M.,Emanuele, S., D�Anneo, A., Vassallo, B., Tesoriere, G. andVento, R. (2003) The effect of 3-aminobenzamide, inhibitor ofpoly(ADP-ribose) polymerase, on human osteosarcoma cells. Int.J. Oncol. 23, 1521–1528.

[9] Rabinovich, G.A., Alonso, C.R., Sotomayor, C.E., Durand, S.,Bocco, J.L. and Riera, C.M. (2000) Molecular mechanismsimplicated in galectin-1-induced apoptosis: activation of the AP-1 transcription factor and downregulation of Bcl-2. Cell DeathDiffer. 7, 747–753.

[10] Sherr, C. (2004) Principles of tumor suppression. Cell 116, 235–246.

[11] Coqueret, O. (2002) Linking cyclins to transcriptional control.Gene 299, 35–55.

[12] Sato, Y., Kobayashi, Y., Sasaki, H., Toyama, T., Kondo, S.,Kiriyama, M. and Fujii, Y. (2003) Expression of Id2 mRNA inneuroblastoma and normal ganglion. Eur. J. Surg. Oncol. 29,284–287.

[13] Gumireddy, K., Ikegaki, N., Phillips, P.C., Sutton, L.N. andReddy, C.D. (2003) Effect of 20-epi-1alpha,25-dihydroxyvitaminD3 on the proliferation of human neuroblastoma: role of cell

cycle regulators and the Myc-Id2 pathway. Biochem. Pharmacol.65, 1943–1955.

[14] Weber, G.F., Ashkar, S., Glimcher, M.J. and Cantor, H. (1996)Receptor-ligand interaction between CD44 and osteopontin (Eta-1). Science 271, 509–512.

[15] Delannoy, P., Lemonnier, J., Hay, E., Modrowski, D. and Marie,P.J. (2001) Protein kinase C- dependent upregulation of N-cadherin expression by phorbol ester in human calvaria osteo-blasts. Exper. Cell Res. 269, 154–161.

[16] Lasorella, A., Iavarone, A. and Israel, M.A. (1996) Id2 specifi-cally alters regulation of the cell cycle by tumor suppressorproteins. Mol. Cell Biol. 16, 2570–2578.

[17] Prabhu, S., Ignatova, A., Park, S.T. and Sun, X.H. (1997)Regulation of the expression of cyclin-dependent kinase inhibitorp21 by E2A and Id proteins. Mol. Cell Biol. 17, 5888–5896.

[18] Strahl, B.D. and Allis, C.D. (2000) The language of covalenthistone modifications. Nature 403, 41–45.

[19] Nguewa, P.A., Fuertes, M.A., Valladares, B., Alonso, C. andPerez, J.M. (2005) Poly(ADP-Ribose)Polymerases: Homology,Structural Domains and Functions. Novel Therapeutical Appli-cations. Prog. Biophys. Mol. Biol. 88, 143–172.

[20] Munoz-Gamez, J.A., Martin-Oliva, D., Aguilar-Quesada, R.,Canuelo, A., Nunez, M.I., Valenzuela, T., De Murcia, G., RuizDe Almodovar, M. and Oliver, F.J. (2004) PARP inhibitionsensitizes p53-deficient breast cancer cells to doxorubicin-inducedapoptosis. Biochem. J. (Sep.), 29.

[21] Manara, M.C., Serra, M., Benini, S., Picci, P. and Scotlandi, K.(2004) Effectiveness of Type I interferons in the treatment ofmultidrug resistant osteosarcoma cells. Int. J. Oncol. 24, 365–372.