University of Zurich - zora.uzh.ch fileMaria Malanga and Felix R. Althaus Institute of Pharmacology and Toxicology, University of Zurich, Winterthurerstrasse 260, CH-8057 Zurich, Switzerland

University of ZurichZurich Open Repository and Archive

Winterthurerstr. 190

CH-8057 Zurich

http://www.zora.uzh.ch

Year: 2004

Poly (ADP-ribose) reactivates stalled DNA topoisomerase I andinduces DNA strand break resealing

Malanga, M; Althaus, F R

Malanga, M; Althaus, F R (2004). Poly (ADP-ribose) reactivates stalled DNA topoisomerase I and induces DNAstrand break resealing. Journal of Biological Chemistry, 279(7):5244-5248.Postprint available at:http://www.zora.uzh.ch

Posted at the Zurich Open Repository and Archive, University of Zurich.http://www.zora.uzh.ch

Originally published at:Journal of Biological Chemistry 2004, 279(7):5244-5248.

Malanga, M; Althaus, F R (2004). Poly (ADP-ribose) reactivates stalled DNA topoisomerase I and induces DNAstrand break resealing. Journal of Biological Chemistry, 279(7):5244-5248.Postprint available at:http://www.zora.uzh.ch

Posted at the Zurich Open Repository and Archive, University of Zurich.http://www.zora.uzh.ch

Originally published at:Journal of Biological Chemistry 2004, 279(7):5244-5248.

Poly (ADP-ribose) reactivates stalled DNA topoisomerase I andinduces DNA strand break resealing

Abstract

Regulating the topological state of DNA is a vital function of the enzyme DNA topoisomerase I.However, when acting on damaged DNA, topoisomerase I may get trapped in a covalent complex withnicked DNA (stalled topoisomerase I), that, if unrepaired, may lead to genomic instability or cell death.Here we show that ADP-ribose polymers target specific domains of topoisomerase I and reprogram theenzyme to remove itself from cleaved DNA and close the resulting gap. Two members of thepoly(ADP-ribose) polymerase family, PARP-1 and 2, act as poly(ADP-ribose) carriers to stalledtopoisomerase I sites and induce efficient repair of enzyme-associated DNA strand breaks. Thus, bycounteracting topoisomerase I-induced DNA damage, PARP-1 and PARP-2 act as positive regulators ofgenomic stability in eukaryotic cells.

1

Poly(ADP-ribose) reactivates stalled DNA topoisomerase I and induces DNA strand break

resealing

Maria Malanga and Felix R. Althaus

Institute of Pharmacology and Toxicology, University of Zurich, Winterthurerstrasse 260, CH-

8057 Zurich, Switzerland

Corresponding author: Felix R. Althaus

Institute of Pharmacology and Toxicology, University of Zurich

Winterthurerstrasse 260, CH-8057 Zurich, Switzerland

Tel: ++41 01 6358762

Fax: ++41 01 6358910

([email protected])

Running title: Poly(ADP-ribose) reactivates stalled DNA topoisomerase I

JBC Papers in Press. Published on December 29, 2003 as Manuscript C300437200

Copyright 2003 by The American Society for Biochemistry and Molecular Biology, Inc.

at Hauptbibliothek U

niversitaet Zuerich Irchel. B

ereich Forschung, on F

ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

2

SUMMARY

Regulating the topological state of DNA is a vital function of the enzyme DNA

topoisomerase I. However, when acting on damaged DNA, topoisomerase I may get trapped in a

covalent complex with nicked DNA (stalled topoisomerase I), that, if unrepaired, may lead to

genomic instability or cell death. Here we show that ADP-ribose polymers target specific

domains of topoisomerase I and reprogram the enzyme to remove itself from cleaved DNA and

close the resulting gap. Two members of the poly(ADP-ribose) polymerase family, PARP-1 and

2, act as poly (ADP-ribose) carriers to stalled topoisomerase I sites and induce efficient repair of

enzyme-associated DNA strand breaks. Thus, by counteracting topoisomerase I-induced DNA

damage, PARP-1 and PARP-2 act as positive regulators of genomic stability in eukaryotic cells.




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

3

INTRODUCTION

DNA topoisomerase I (topo I) plays an essential role in controlling the level of DNA

supercoiling and relieving the torsional stress that is generated during replication, transcription,

recombination and chromatin remodeling (1,2). However, when scanning the topological state of

DNA, topo I may get trapped in the vicinity of DNA lesions and cause secondary DNA damage

by an abortive mechanism (3). This involves the formation of a DNA single strand break,

covalent attachment of the enzyme to the 3’-phosphate terminus (cleavage complex), and

inhibition of the DNA ligase activity of topo I (“stalled topo I”). Stalled topo I is a threat to

genomic stability and cell life since enzyme-bound nicked DNA can be converted into

irreversible single- or double-strand breaks. In fact, the efficacy of camptothecins, a prominent

class of anticancer drugs (also known as topo I poisons) is due to the stabilization of topo I

cleavage complexes (4). In eukaryotes, a specific repair enzyme, tyrosyl-DNA phosphodiesterase

1 (TDP1) has evolved to counteract the secondary DNA damage of stalled topo I by removing

the enzyme from the 3’-terminus. Nevertheless, a DNA single strand break is generated that

needs to be repaired by the base excision repair machinery (5,6).

Here we show that two members of the poly(ADP-ribose) polymerase family, i.e. PARP-

1 and PARP-2 can remove stalled topo I from DNA. PARP-1 and PARP-2 belong to the DNA

damage surveillance network (7,8). They are activated by DNA strand breaks to catalyze the

conversion of the respiratory coenzyme nicotinamide adenine dinucleotide (NAD+) into protein-

bound ADP-ribose polymers, with concomitant release of nicotinamide (9). While PARP-1 has

been found to poly (ADP-ribosyl)ate a number of nuclear proteins (“heteromodification”) (9

and references therein), both PARP-1 and PARP-2 preferentially modify themselves




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

4

(“automodification”) (10, 11). We found that, with the attached polymers, PARP-1 and PARP-2

bind to specific domains of topo 1 and reprogram their functions. Even camptothecin-stabilized

cleavage complexes, resembling those formed by topo I in the vicinity of DNA lesions (3), could

be disjoined by this mechanism.




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

5

Materials and Methods

Enzymes and oligonucleotides

Human DNA topoisomerase I, certified to be purified to homogeneity and free of nuclease and

topoisomerase II contaminations, was purchased from TopoGen, together with camptothecin and

polyclonal anti-human topoisomerase I antibodies.

Recombinant PARP-1 and PARP-2 (Alexis), [γ-32P]-ATP (7000 Ci/mmol) (ICN), and the

oligonucleotides (Microsynth) 5’-CTAGTAAAAGACTTGGAAAAATTTTTAAAAAA-3’, 5’-

AATTTTTTTTAAAAATTTTTCCAAGTCTTTTA-3’; 5’-CTAGTAAAA GACTTGGA-3’; 5’-

TTTTTTAAAAATTTTTCCAAGTCTTTTACTAG-p were of the highest purity commercially

available. The sequences of the oligonucleotides are essentially as reported (12,13). 5’-end

labeling of the scissile strand oligonucleotide was performed using T4 polynucleotide kinase and

[γ-32P]-ATP, annealed to the unlabeled complementary strand, and purified by polyacrylamide

gel eletrophoresis. The site of topo I cleavage is indicated by an arrow in Figures 1 and 2.

Automodified PARPs and poly(ADP-ribose)

In the automodification reaction , PARP-1 or PARP-2 were incubated for 30 min with 0.1-0.4

mM [α-32P]-NAD (4000 dpm/nmol), as described by Beneke et al. (14). Mock automodification

was carried out in the presence of 2 mM 3-aminobenzamide (AB). The products were purified by

ultrafiltration on Microcon YM-30 membrane. Automodified or mock modified enzymes were

used freshly for DNA binding/cleavage assays or stored at –80 °C in 50 mM Tris-HCl buffer,




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

6

pH 7.4, containing 14 mM β-Mercaptoethanol, 0.5 mM PMSF, 20% glycerol. Free poly(ADP-

ribose) was prepared as previously described (15) and their size distribution determined (16).

Topo I reactions

Purified human topoisomerase I (0.5 pmol) was incubated with [32P]-5'-end labeled

oligonucleotide substrates (20’000-40’000 dpm; 20 fmoles) in 50 mM Tris-HCl, pH 8.0

including 0.1 mg/ml BSA either at 37 °C for 10 min or at 30 °C for 15 min (suicide substrate).

Reactions were carried out in 20 µl (final volume) and stopped by adding 0.5 volumes of 3x

Laemmli loading buffer. Covalent complexes of topoisomerase I with cleaved oligonucleotides

were separated from the uncleaved substrates by 10% SDS-PAGE and visualized by

autoradiography. Camptothecin (CPT) was used at 20 µM concentration. Autoradiographic

bands were quantified by scanning densitometry (Molecular Dynamics); for a representative set

of samples, the densitometric quantification of enzyme-bound and free DNA was paralleled by

liquid scintillation counting of excised radioactive bands. The fraction of the oligonucleotide

substrate in the cleavage complex ranged between 2-5% and 10-25%, in the absence or presence

of CPT, respectively.

Poly(ADP-ribose), either PARP-bound or as free polymer, was present in the reaction

mixtures at the amounts indicated in the Figures (pmol quantities refer to the ADP-ribose unit);

reactions were started by the addition of the radioalabeled oligonucleotide substrates. For

reversal experiments, topo I - DNA complexes were allowed to form over 10 min incubation in

the conditions described above, then either poly(ADP-ribose) or 0.35 M NaCl (12,17) were

added and incubation was extended for the time indicated in the figure legends. After SDS-

PAGE and autoradiography, remaining cleavage complexes were quantified by scanning




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

7

densitometry. For reversal kinetics, the complex formed before any addition was set at 100% and

percentage data were plotted on a semilogarithmic scale against time. The religation constant (kr)

was calculated by fitting these data to the equation Ct = C0e-krt (where Ct = % remaining cleavage

complex at the end of the incubation time, t; C0 = % cleavage complex at t0 = 100). Kr values

were converted to half life (t1/2) values according to the equation t1/2 = ln 2/kr (18).

Binding of topo I to dsDNA was determined by electrophoretic mobility shift (19). Topo

I (45 ng) was incubated with 20 fmol of radiolabeled oligonucleotide, in the conditions described

above. At the end of the incubation time, half the reaction mixtures (10 µl) were withdrawn,

mixed with 0.25 volumes of loading buffer (10 mM Tris-HCl, pH 8.0, 15% Ficoll, 0.05%

Bromophenol Blue) and electrophoresed on native 6% polyacrylamide gel. The remaining half of

the reactions was analysed by SDS-PAGE to evaluate the potential contribution of covalent

complexes to the mobility shift. In the conditions used to visualize the noncovalent topo I-DNA

complex, the cleavage conjugate was undetectable.

Poly (ADP-ribose) binding assay

The noncovalent binding of poly(ADP-ribose) to topo I (and other proteins) was determined as

previously described (20). Purified proteins were either dot blotted or transferred onto

nitrocellulose membrane after electrophoretic separation on 10% SDS-polyacrylamide gel.

Bound poly(ADP-ribose) was visualized by autoradiography and quantified by scintillation

counting of excised radioactive dots. Western blotted proteins were gold stained using the

Protogold staining kit (British BioCell International), as recommended by the manufacturer.




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

8

RESULTS AND DISCUSSION

Poly (ADP-ribose) regulates the formation of topo I cleavage complexes

The data in Fig. 1A show that PARP-1 and PARP-2 block the formation of topo I covalent

complex with DNA only in the poly ADP-ribosylated form, while the unmodified enzymes,

incubated under identical conditions with the inhibitor 3-aminobenzamide (AB), were

completely ineffective. Moreover, PARP-1 and 2 were still effective, when topo I complex

formation was strongly enhanced in the presence of camptothecin (CPT, Fig. 1A, bottom), but

again only after the enzymes had been automodified with ADP-ribose polymers. These data

suggested that the ADP-ribose polymers attached to PARP-1 or PARP-2 could be responsible for

the inhibitory effects on topo I complex formation. Direct evidence for this is shown in Fig. 1B.

Free ADP-ribose polymers detached from PARP-1 or PARP-2 expressed the full inhibitory

activity in a dose-dependent manner (Fig. 1B), while a 4-fold excess (w:w) of a 39-nucleotide

single stranded DNA did not influence the reaction (Fig. 1C, right panel). Degradation of

polymers to monomeric ADP-ribose, using poly(ADP-ribose)glycohydrolase (PARG),

completely abolished the inhibitory principle (Fig. 1C, left panel).

Poly(ADP-ribose) could prevent the initial DNA binding of topo I which then would

preclude subsequent formation of covalent complexes. Fig. 2A shows that DNA binding, as

measured by electrophoretic mobility shift analysis, was not affected by ADP-ribose polymers at

quantities up to 20 pmol that completely blocked covalent complex formation (Fig. 1). Also,

(ADP-ribose) polymers did not accelerate dissociation of topo I already bound to DNA after a 10

min preincubation (Fig. 2A, lane 5). On the other hand, 0.35 M NaCl was sufficient to prevent




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

9

formation of topo I - DNA noncovalent complexes as well as to dissociate them (Fig. 2A, lanes 4

& 6). It should be mentioned that in these types of analyses (see also Fig. 2D) the shifted topo I –

DNA complexes always migrated as two discrete bands. The high purity of the topo I

preparations (see “Materials and Methods”) and the fact that both bands were reduced when

incubation was carried on in the presence of anti-topo I antibodies, but not in the presence of

non-specific immunoglobulins (data not shown), argue against DNA binding by protein

contaminants. It seems also unlikely that the two complexes result from DNA binding of

differently sized topo I fragments. In fact, in western blot analyses, a single component of the

expected molecular weight was detected by polyclonal anti-topo I antibodies (Fig. 2A, right

panel); in addition, topo I, either labeled by covalent binding to radiolabeled oligonucleotides

(Fig. 1-3) or by noncovalent interaction with [32P]-poly (ADP-ribose) (Fig. 4), migrated as a

single band in SDS-polyacrylamide gels. Thus, generation of different conformers of topo I-

DNA complexes as well as binding of one or two topo I molecules/oliglonucleotide molecule

might underlie the peculiar electrophoretic behaviour under native conditions. On the other hand,

topo I dimers have been shown to form both in vitro (21) and in vivo (22).

Poly (ADP-ribose) affects both the forward and reverse DNA transesterification reaction of

topo I, but with opposite effects

The reduced yield of cleavage complexes in the presence of poly (ADP-ribose) could be

explained by either inhibition of DNA cleavage (forward transesterification reaction) or by a

shift in the cleavage/religation equilibrium in favour of religation (reverse transesterification

reaction). To examine whether poly (ADP-ribose) could affect DNA cleavage, the

oligonucleotide shown in Fig. 2B was used as a suicide substrate for topo I: in fact the enzyme




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

10

becomes trapped in the covalent complex (suicide complex), since the cleavage reaction

generates a free trinucleotide with the 5’-OH end that is lost by diffusion and hence religation

becomes impossible (12). The results of Fig. 2B demonstrate that poly(ADP-ribose) is a potent

inhibitor of DNA cleavage by topo I: as little as 1 pmol of polymeric ADP-ribose was sufficient

for complete inhibition (Fig. 2B). Conversely, poly (ADP-ribose) had no effect on preformed

suicide complexes (Fig. 2B, compare lanes 1 & 6). Again, inhibition of DNA cleavage was also

achieved with poly(ADP-ribosyl)ated PARP-1 and PARP-2 (Fig. 2C & D), but not with the

unmodified enzymes (Fig. 2C, lanes 4 & 5). Incidentally, noncovalent topo I binding to the

suicide substrate was equally unaffected by poly(ADP-ribosyl)ated PARP-1 and PARP-2 (Fig.

2D), as has been observed with the reversible cleavage substrate (Fig.2A).

Next we determined whether poly(ADP-ribose) could reverse preformed covalent topo I-

DNA conjugates by stimulating the religation activity of topo I. In the experiment shown in Fig.

3A, complex formation was allowed to proceed for 10 and 20 min in the absence or presence of

camptothecin. Under these conditions, a steady state level of cleavage complexes was reached

within 10 min. Following addition of ADP-ribose polymers, the topo I – DNA complexes

disappeared (Fig. 3A, lane 3). Strikingly, this was also observed with the complexes stabilized

in the presence of camptothecin (Fig. 3A, lane 6), indicating that poly (ADP-ribose) is able to

overcome the poisoning effect of the drug. Removal of cleavage complexes was achieved with

poly (ADP-ribosyl)ated PARP-1 (Fig. 3B) or PARP-2 (not shown), but not with the unmodified

enzymes (Fig. 3B, lane 6), emphasizing that polymers are essential for this effect. It should be

noted that due to the strong inhibition exerted by poly (ADP-ribose) on the DNA cleavage

reaction (Fig. 2B-D), the disappearance of cleavage complexes in these reversal experiments was

essentially dependent on the rate of religation. Uncoupling of cleavage and religation reactions




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

11

can also be achieved by increasing salt concentration. In fact 0.35 M NaCl blocks cleavage,

probably by interfering with de novo binding of the enzyme to DNA (12,17, and Fig 2A), but has

no effect on the DNA ligation activity of topo I. The effect of CPT is also reversed under such

conditions (17). A comparison of the religation kinetics in the presence of either poly (ADP-

ribose) or 0.35 M NaCl indicated that DNA strand break resealing by topo I is indeed accelerated

by poly (ADP-ribose) (Fig. 3C & D): the half life of cleavage complexes was more than 2 fold

shorter in the presence of poly (ADP-ribose). Taken together, these results suggest that the low

yield of cleavage complexes (Fig. 1) and the efficient reversal of such complexes (Fig. 3) may

arise from two opposing actions of poly(ADP-ribose): inhibition of DNA cleavage (Fig. 2) and

stimulation of the topo I religation activity (Fig. 3).

Poly (ADP-ribose) binds noncovalently to topo I

With their long polymeric arms, members of the PARP family can reach out to bind specific

proteins (23-25) and alter their domain functions (20,26). Several DNA damage checkpoint

proteins are targeted in this manner and the mechanism involves one or several poly(ADP-

ribose)-binding sequence motifs in the target protein (27). The motif, a 20 to 25 amino acid

sequence with regularly spaced hydrophobic residues, interspersed with basic residues and

flanking arginines and lysines, was found to occur at three conserved sites in the topo I sequence

(Fig. 4A). These sites are located in structurally and functionally distinct protein domains (28):

two sequences, comprising amino acids 261-280 and 532-550, are in the core DNA binding

domain, but at opposite sites of the clamp formed by the enzyme around DNA; a third polymer-

binding site (sequence 669-688) is in the linker domain. The latter connects the DNA binding

and C-terminal domains and has a regulatory role on the overall reaction as it slows down the




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

12

religation step and thereby facilitates DNA relaxation (18,29). Biochemical analysis using a

nitrocellulose blot assay (20) directly confirmed that topo I belongs to the family of poly(ADP-

ribose)-binding proteins (Figs. 4B & C). In the dot blot assay (Fig. 4B), about 1 pmol polymeric

ADP-ribose/pmol protein bound to topo I and histone H1 (positive control) immobilized on

nitrocellulose. Poly (ADP-ribose) binding also occurred after electrophoretic separation and

western blotting of these proteins (Fig. 4C).

Mapping the poly (ADP-ribose)-binding sites into the published crystal structure of

human topo I (28) provided a plausible explanation for the differential effects of ADP-ribose

polymers on specific domain functions. First, the polymer-binding regions have no primary role

in the DNA-binding of topo I, which may explain why poly (ADP-ribose, while interacting with

the free enzyme (Fig. 4B &C) does not affect topo I - DNA noncovalent complexes (Fig. 2A &

D). On the other hand, the sequence comprising amino acids 532-550 is located into core sub-

domain III that largely contributes to the formation of the catalytic site: binding of poly (ADP-

ribose) at this site may underlie the strong inhibition of the DNA cleavage reaction. The cap

region of topo I (residues 175-433), which includes the poly (ADP-ribose)-binding sequence

261-280, is also essential for DNA cleavage (30).

How poly (ADP-ribose) counteracts CPT action and accelerates DNA religation is less obvious

and requires further considerations. The x-ray crystal structures of human topo I in covalent

complex with DNA in the absence and presence of a CPT analogue, topotecan, have revealed the

mode of binding of the drug and clarified important mechanistic aspects of topo I poisoning (31).

Topotecan binds to the DNA-enzyme complex by intercalating into DNA at the site of cleavage

and displacing the nucleophilic 5’-OH end of the cleaved strand: thus religation is prevented.

The comparison of the structure of the ternary (enzyme - DNA - drug) and binary (enzyme –




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

13

DNA) complexes also showed that the most dramatic changes involved the linker domain, which

could be visualized in the electron density map only after drug binding. Interestingly, deletion of

the linker domain, either total (18) or partial (32), or even a single amino acid substitution (33)

make the enzyme insensitive to CPT. Alltogether, these observations point to an involvement of

this domain in the stabilization of the ternary complex. It should be noted that the linker-

associated poly (ADP-ribose)-binding site (amino acids 669-688) is likely to be the most (if not

the only) accessible one to the polymer when topo I is engaged in the cleavage complex. Thus,

we speculate that targeting of the linker domain by poly (ADP-ribose) may have a dual effect: i)

it may induce conformational changes that destabilize the CPT binding pocket and reposition the

broken DNA strand so that ligation can occur (neutralization of CPT poisoning; Fig. 3); ii) it

may alleviate the restrains imposed by the linker domain on DNA religation (18,29), which

would explain the reduced half life of cleavage complexes (Fig. 3C & D).

In conclusion, both PARP-1 and PARP-2 are able to disjoin covalent DNA complexes

and induce the reversal of DNA breaks that are formed when topo I gets stalled in the vicinity of

DNA lesions. DNA breaks are also the activating signal for PARP-1 and PARP-2 (7-11) to

produce (ADP-ribose)n-polymers which we show here to be the molecules responsible for

removing topo I from DNA. Thus, an important aspect of PARP-1 and PARP-2 function in

preserving genomic stability of cells is to deploy polymers to sites of stalled topo I. By contrast

to the topo I removal mechanism catalyzed by tyrosyl-DNA phosphodiesterase 1 (5,6), this

PARP-based rescue mechanism has an immediate impact on the genomic stability of cells, as it

is activated within seconds and does not leave an unfilled DNA gap.




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

14

Our data also bear on the design of new therapeutic strategies for the treatment of

cancers. As shown in Figs. 1 & 3, poly(ADP-ribose) may counteract the effects of the cancer

therapeutic drug camptothecin and contribute to drug resistance. Thus, the efficacy of

camptothecin-mimetic drugs can be enhanced by combination with PARP inhibitors. The in vivo

relevance supporting this concept has already been demonstrated: cancer cells with compromised

PARP-1 function are hypersensitive to camptothecin (34) and the cytotoxicity of topo I-targeted

drugs is enhanced by PARP inhibitors (34,35). Finally, the differential effects of poly(ADP-

ribose) on particular domain functions of topo I, i.e. the DNA cleavage and ligation reactions can

be exploited by site-specific targeting and this opens new perspectives for the design of novel

topo I – based anticancer drugs.




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

15

REFERENCES

1. Wang , J. C. (2002)Nature Rev. Mol. Cell. Biol. 3, 430-440.

2. Postow, L., Crisona, N. J., Peter, B. J., Hardy, C. D. & Cozzarelli, N.R. (2001) Proc. Natl.

Acad. Sci. USA 98, 8219-8226.

3. Pourquier, P. & Pommier Y. (2001) Adv. Cancer Res. 80, 189-216.

4. Li, T.-K & Liu, L. F. (2001) Annual Rev. Pharmacol. Toxicol. 41, 53-77.

5. Yang, S.-W., Burgin, A. B. Jr, Huizenga, B. N., Robertson, C. A., Yao, K. C., Nash, H. A.

(1996) Proc. Natl. Acad. Sci. USA 93, 11534-11539.

6. Caldecott, K. W. (2003) Cell 112, 7-10.

7. Schreiber, V., Amé, J. C., Dolle, P., Schultz I., Rinaldi, B., Fraulob, V., Menissier-de Murcia,

J. & de Murcia, G. (2002) J. Biol. Chem. 277, 23028-23036.




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

16

8. Menissier de Murcia, J., Ricoul, M., Tartier, L., Niedergang, C., Huber, A., Dantzer, F.,

Schreiber, V., Amé, J. C., Dierich, A., LeMeur, M. et al. (2003) EMBO J. 22, 2255-2263.

9. D’Amours, D., Desnoyers, S., D’Silva, I. & Poirier, G. G. (1999) Biochem. J. 342, 249-268.

10. Ogata, N., Ueda, K., Kawaichi, M. & Hayaishi, O. (1981) J. Biol. Chem. 256, 4135-4137.

11. Amé, J.-C., Rolli, V., Schreiber, V., Niedergang, C., Apiou, F., Decker, P., Muller, S., Hoger,

T., Ménissier-de Murcia, J. & de Murcia, G. (1999) J. Biol. Chem. 274, 17860-17868.

12. Pourquier, P., Ueng, L. M., Fertala, J., Wang, D., Park, H. J., Essigmann, J. M., Bjornsti, M.

A. & Pommier, Y. (1999) J. Biol. Chem. 274, 8516-8523.

13. Pourquier, P., Ueng, L. M., Kohlhagen, G., Mazunder, A., Gupta, M., Kohn, K. W. &

Pommier, Y. (1997) J. Biol. Chem. 272, 7792-7796.

14. Beneke, S., Alvarez-Gonzalez, R. & Burkle, A. (2000) Exp. Gerontol. 35, 989-1002.

15. Malanga, M., Bachmann, S., Panzeter, P. L., Zweifel, B. & Althaus, F. R. (1995) Anal.

Biochem. 228, 245-251.

16. Panzeter, P. L. & Althaus, F. R. (1990) Nucleic Acids Res. 18, 2194.




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

17

17. Tanizawa, A., Kohn, K. W., Kohlhagen, G., Leteurtre F. & Pommier, Y. (1995) Biochem.

34, 7200-7206.

18. Stewart, L., Ireton, G. C. & Champoux, J. J. (1999) J. Biol. Chem. 274, 32950-32960.

19. Shuman, S. (2001) Methods Mol. Biol. 95, 65-74.

20. Panzeter. P. L., Zweifel, B., Malanga, M., Waser, S. H., Richard, M. & Althaus, F. R. (1993)

J. Biol. Chem. 268, 1762-1764.

21. Soe, K., Dianov, G., Nasheuer, H.-P., Bohr, V. A., Grosse, F. & Stevnsner T. (2001), Nucleic

Acids Res. 29, 3195-3203.

22. Mao, Y., Okada, S., Chang, L.-S. & Muller, M. (2000) Cancer Res. 60, 4538-4543.

23. Realini, C. & Althaus, F.R. (1992) J. Biol. Chem. 267, 18858-18865.

24. Okano, S., Lan, Li, Caldecott , K. W., Mori, T. & Yasui, A. (2003) Mol. Cell. Biol. 23, 3974-

3981.

25. Gagné, J. P., Hunter, J. M., Labrecque, B., Chabot, B. & Poirier, G. G. (2003) Biochem. J.

371, 331-340.




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

18

26. Malanga, M., Pleschke, J.M., Kleczkowska & Althaus, F. R. (1998) J. Biol. Chem. 273,

11839-11843.

27. Pleschke, J.M., Kleczkowska, H. E., Strohm, M. & Althaus, F. R (2000) J. Biol. Chem. 275,

40974-40980.

28. Redinbo, M. R., Stewart, L., Kuhn, P., Champoux J. J.& Hol, W.G. J. (1998) Science 279,

1504-1513.

29. Stewart, L., Redinbo, M. R., Qiu, X., Hol, W. G. J. & Champoux, J. J. (1998) Science 279,

1534-1541.

30. Yang, Z. & Champoux, J. J. (2002) J. Biol. Chem. 277, 30815-30823.

31. Staker, B. L., Hjerrild, K., Feese, M. D., Behnke, C. A., Burgin, A. B., Jr. and Stewart, L.

(2003) Proc. Natl. Acad. Sci. USA 99, 15387-15392.

32. Ireton, G. C., Stewart, L., Parker, L. H. & Champoux, J. J. (2000) J. Biol. Chem. 275, 25820-

25830.

33. Fiorani, P., Bruselles, A., Falconi, M., Chillemi, G., Desideri, A. & Benedetti, P. (2003) J.

Biol. Chem. 278, 43268-43275.




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

19

34. Chatterjee, S., Cheng, M.F., Trivedi, D., Petzold, S. J. & Berger, N. A. (1989) Cancer

Commun. 1, 389-394.

35. Bowman, K. J., Newell, D. R., Calvert, A. H. & Curtin, N. J. (2001) Br. J. Cancer 84, 106-

112.

36. Miknyoczki, S. J., Jones-Bolin, S., Pritchard, S., Hunter, K., Zhao, H., Wan, W., Ator, M.,

Bihovsky, R., Hudkin, R., Chatterjee, S. et al. (2003) Mol. Cancer Ther. 2, 371-382.

Acknowledgments

This work was supported by the Swiss National Science Foundation.




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

20

Figure legends

Figure 1 – Poly(ADP-ribose), either PARP-1 or PARP-2 bound, or as a free polymer, regulates

the level of cleavage complexes formed on a topo I substrate (top). 1A, complex formation in the

absence or presence of either automodified (lanes 4 and 7) or mock poly (ADP-ribosyl)ated

PARPs (PARP/AB, lanes 5 and 6). PARP-1*, PARP-2* (20 pmol enzyme-bound poly(ADP-

ribose). Lanes 2 and 3, native PARP-2 (250 and 500 fmol, respectively). 1B, cleavage complexes

(mean +/- S.D., n ≥ 4) in the presence of increasing amounts (pmol) of poly(ADP-ribose) and

SDS-PAGE analysis (insert). 1C, left, effect of either intact (lanes 3 and 5) or PARG-digested

(lanes 2 and 4) poly(ADP-ribose); right, effect of a 39mer ss-oligonucleotide.

Figure 2– Poly(ADP-ribose) does not interfere with topo I binding to, or dissociation from

DNA, but it inhibits DNA cleavage. 2A, left: electrophoretic mobility shift analysis of topo I

DNA-binding in the absence (lane 2) or presence of either poly (ADP-ribose) (20 pmol; lane 3)

or 0.35 M NaCl (lane 4). Alternatively, buffer (lane 7), poly (ADP-ribose) (20 pmol; lane 5) or

NaCl (0.35 M; lane 6) were added after 10 min preincubation of topo I with the radiolabeled

oligonucleotide and reactions were continued for 10 min further. About 4% of input DNA

(sequence shown in Fig. 1A) was shifted by topo I. Lane 1: topo I omitted. 2A, right: western

blot analysis. Topo I binding by specific polyclonal antibodies was detected by enhanced

chemiluminescence (ECL). 2B, lanes 1 – 4, 15 min incubation of suicide substrate (sequence

shown on top) with topo I +/- the indicated amounts of poly(ADP-ribose); lanes 5 and 6:

incubation was for 30 min: buffer (lane 5) or 5 pmol poly(ADP-ribose) (lane 6) were added after

the first 15 min of incubation. 2C, reaction was carried on in the absence (lane 1) or presence of




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

21

either native (lanes 4,5) or poly(ADP-ribosylated) PARPs (*PARP: 1 pmol PARP-bound

polymeric ADP-ribose). 2D, native PAGE analysis of topo I noncovalent and covalent (suicide)

complexes (*PARP: 20 pmol PARP-bound polymeric ADP-ribose). The fraction of input DNA

in the suicide complex increased with incubation time and was 7-10% after 15 min.

Figure 3 – Poly(ADP-ribose) reverses preformed topo I DNA cleavage complexes. 3A, cleavage

complexes were allowed to form in 10 min incubation (lanes 1 and 4), then either buffer (lanes 2

and 5) or 20 pmol poly(ADP-ribose) (lanes 3 and 6) were added and incubation was continued

for additional 10 min. 3B, time course of the disappearance of CPT-stabilized cleavage

complexes upon addition of 20 pmol PARP-bound poly(ADP-ribose) (PARP-1*); lane 1:

preformed complexes; lanes 2 – 4: cleavage complexes 1, 2 and 5 min after PARP-1* addition.

Buffer (lane 5) or native PARP-1 (lanes 6) served as controls. 3C, comparison of the kinetics of

cleavage complex reversal in the presence of either poly (ADP-ribose) (20 pmol) or NaCl.

Following additions, incubations were continued for 1 (lanes 2 and 5), 2 (lanes 3 and 6) or 5 min

(lanes 4 and 7). 3D, semilogarithmic plot of data from 3C. Remaining cleavage complexes are

expressed as percentage of the complexes formed during 10 min incubation, before any addition

(lane 1 in 3C, control). Reactions were carried on in the presence of CPT.

Figure 4 – Poly(ADP-ribose) binds noncovalently to topo I. 4A, top: sequence alignments with

a poly (ADP-ribose)-binding motif. For each peptide, the positions of the first and last amino

acids in the human topo I sequence are indicated. 4A, bottom: domain structure of topo I (18);

white boxes with vertical dotted lines locate the poly (ADP-ribose)-binding sites. 4B, [32P]-poly




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

22

(ADP-ribose) probing of proteins dot-blotted on nitrocellulose membrane; 4C, proteins were

electrophoretically separated before transfer on nitrocelluose and poly (ADP-ribose) binding

assay. Left panel: autoradiogram; right panel: gold staining of proteins on nitrocellulose. H: 1 µg

calf thymus histone H1; M: molecular weight markers, Precision Plus, BioRad; T: 0.2 µg

purified topo I.




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

A

1 2 3 4 5 6 7

No additi

on

PARP-2

PARP-2

PARP-2*

PARP-1*

PARP-2/A

B

PARP-1/A

B

- CPT

+ CPT

AT T T T C T G AACC T T T T T AAAAAT T T T T T T T AA-5’3’-C T AG T AAAAG AC T T GG AAAAAT T T T T AAAAAA-3’5’-[ P]-32

C

+ CPT

PARG + + - + -

- CPT+ CPT1 2 3 4 5 6 7

no additi

on

ss-3

9mer

B

0 5 10 20 0 5 10 20

- CPT + CPT

Poly(ADP-ribose)(ADP-ribose pmol)

20

40

60

80

100

0 5 10 15 20clea

vag

e co

mp

lex

(% o

f co

ntr

ol)

Fig. 1




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

B

C

-3’C T A G T A A A A G A C T T G G A5’-[ P]-32

-5’-P3’-G A T C A T T T T C T G A A C C T T T T T A A A A A T T T T T T

1 2 3 4 5 6

0 1 2.5 5 0 5pmol

1 2 3 4 5

No additi

on

PARP-1

PARP-2

PARP-2*

PARP-1*

D

Fig. 2

A

noncovalentcomplexes

No additi

on

PARP-2*

PARP-1*

free probe

suicidecomplex

1 2 3 4 5 6 7

(ADP-ribose)n - - + - + - -- - - + - + -0.35 M NaCl

ECL




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

Fig. 3

A

(ADP-ribose)n - - + - - +

1 2 3 4 5 6- CPT + CPT

B

PARP-1* - + + + - -

1 2 3 4 5 6

(ADPR)n0.35 M NaCl

- + + + - - -

- - - - + + +

1 2 3 4 5 6 7

C

D

clea

vag

e co

mp

lex

(% o

f co

ntr

ol)

time (min)

100

10

10 1 2 3 4 5 6

+ (ADPR)n

+ NaCl

time (min)




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

Fig. 4

A

hx b bb bx h hh h

K A D A K V M K D A K T K K - V V E S K K

A K M L D H E Y T T K E I F - R K N F F KK D S I R Y Y N K V P V E K R - V F K N

261 280532 550

669 688

B

H1 Prot K DNase I H110 ng 1 µg 1 µg 50 ng

11 22 45 90 topo I (ng)

H M T H M T250150

10075

50

37

C

Core domainN-terminal domain Linker

C-terminal domain

1 215 636 713 765N- -C




ebruary 9, 2010w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org/

Documents

University of Zurich - zora.uzh.ch fileMaria Malanga and Felix R. Althaus Institute of Pharmacology and Toxicology, University of Zurich, Winterthurerstrasse 260, CH-8057 Zurich, Switzerland