DNA recombinase activity of eukaryotic DNA topoisomerase I; effects of camptothecin and other inhibitors

DNA

ELSEVIER

Repair

Mutation Research 337 (1995) 135-145

DNA recombinase activity of eukaryotic DNA topoisomerase effects of camptothecin and other inhibitors Yves Pommier

*,

I;

Jeffrey Jenkins, Glenda Kohlhagen, Frangois Leteurtre

Laboratory of Molecular Pharmacology, Developmental Therapeutics Program, Division of Cancer Treatment, National Cancer Institute, Bldg. 37, Rm. X25, National Institutes of Health, Bethesda, MD 20892-4255, USA

Received 13 February 1995; revised 17 April 1995; accepted 18 April 1995

Abstract

DNA oligonucleotides containing a strong topoisomerase I cleavage site were used to study the DNA cleavage and strand transferase activities of calf thymus topoisomerase I (top11 in the absence and presence of camptothecin. A partially single-stranded oligonucleotide with only two nucleotides on the 3’ side of the cleavage site (positions + 1 and +2) was cleaved at the same position as the corresponding duplex oligonucleotide. However, cleavage in the absence of camptothecin was more pronounced than in the duplex oligonucleotide and was only partially reversible in the presence of 0.5 M NaCl, consistent with release of the dinucleotide 3’ to the top1 break. Another reaction took place generating a larger DNA fragment which resulted from religation (strand transfer) of the S-hydroxyl terminus of the non-scissile DNA strand to the 3’ end of the topl-linked oligonucleotide after loss of the + 1 and + 2 nucleotides. Top1 religation activity appeared efficient since only the last 5’ base of the single-stranded DNA acceptor was complementary to the 3’ tail of the donor DNA. Religation was not detectable with a double-stranded DNA acceptor, which is consistent with the persistence of topl-induced DNA double-strand breaks in camptothecin-treated cells. Camptothecin and other top1 inhibitors enhanced cleavage in both the partially single-stranded and the duplex oligonucleotides, indicating that they did not inhibit the induction of topl-mediated DNA cleavage but primarily blocked the religation step of the enzyme catalytic cycle. The top1 DNA strand transferase activity was reversibly inhibited by camptothecin and several derivatives, as well as saintopin. These results are discussed in terms of camptothecin-induced DNA recombinations. Keywords:

DNA strand transfer; Cancer chemotherapy;

1. Introduction Mammalian are ubiquitous

type I DNA topoisomerases enzymes that make transient

(topl) DNA

* Corresponding author. Fax (301) 402-0742; E-mail [email protected]. Elsevier Science B.V. 0921-8777(95)00019-4

SSDI

Illegitimate

recombination;

Saintopin

single-strand breaks by transferring a DNA phosphodiester bond to a conserved enzyme tyrosine (presumably Y723 for human topl) Khampoux, 1990; Hsieh, 1992; Pommier and Tanizawa, 1993). These transient topl-linked DNA breaks are commonly referred to as cleavable complexes because they are only detectable after protein denaturation. The top1 covalent linkage is with the

136

Y. Pommier

et al. /Mutation

3’ terminus of the broken DNA while the other terminus is a S-hydroxyl. Cleavable complexes occur readily in duplex DNA with limited base sequence selectivity, except for a preference for T at the 3’ terminus of the break (position - 1 relative to the cleavage site) (Been et al., 1984; Jaxel et al., 19881991; Kjeldsen et al., 1988; Porter and Champoux, 1989). Single-stranded DNA is not cleaved by top1 unless a hairpin loop provides a duplex segment (Been and Champoux, 1984). Top1 cleavable complexes are sites of DNA strand passage (or swivelling) that play an essential role in allowing DNA relaxation and removal of DNA torsional tension during replication, transcription and repair (Champoux, 1990; Pommier and Tanizawa, 1993). Following DNA strand passage, top1 cleavable complexes are reversed by religation of the S-hydroxyl terminus to the 3’-phosphate linked to topl. Camptothecin is a potent anticancer agent and several of its derivatives are in clinical trials with promising results (Pommier and Tanizawa, 1993; Slichenmyer et al., 1993; Potmesil, 1994). Camptothecin produces in cells protein-linked DNA strand breaks that correspond to the stabilization of top1 cleavable complexes (Hsiang et al., 1985; Covey et al., 1989). The cytotoxicity of camptothecin is due to the conversion of a fraction of

A

Research 337 (1995)

135-145

the camptothecin-induced cleavable complexes into DNA double-strand breaks by colliding replication forks (Holm et al., 1989; Hsiang et al., 1989) (see Fig. 1A). DNA double-strand breaks have been detected in replicating simian virus 40 (Snapka and Permana, 1993; Tsao et al., 1993) and mammalian cells (Ryan et al., 1991) treated with camptothecin. Their cytotoxicity may be attributable to slow repair and mutations (Fig. 1C) (Ryan et al,, 1991; Pommier and Tanizawa, 1993). Top1 cleavable complexes can also be converted into topl-linked single-stranded DNA gaps when the cleaved DNA strand 3’ to the top1 break contains a nick less than six bases away from the cleaved DNA phosphodiester bond (Fig. 1B) (Svejstrup et al., 1991; Shuman, 1992a,b; Christiansen et al., 1993). In this case, the DNA strand 3’ to the top1 break can dissociate and top1 loses its religation partner. These irreversible reactions have been referred to as ‘suicide reactions’ (Svejstrup et al., 1991; Shuman, 1992a). Top1 can then promote illegitimate recombination after strand invasion by a single-stranded DNA partially homologous to the leaving strand and bearing a 5’-hydroxyl terminus (Fig. 10 (Halligan et al., 1982; Champoux and Bullock, 1988; Svejstrup et al., 1991; Shuman, 1992a,b; Christiansen et al., 1993).

B ~ r,

3

\rfork ----4 ---*

3’

C Str8nd Invasion

Fig. 1. Camptothecin-induced DNA lesions as a result of collision with a replication fork (A), or presence of nick 3’ of the top1 cleavage site (B). (C) Top1 (hatched circles) can recombine DNA by strand transfer with an invading DNA single-strand bearing a 5’-hydroxyl and partial homology with the displaced strand.

Y. Pommier et al. /Mutation Research 337 (1995) 135-145

A

5, -

33u

GATCTAAAAGACTT-GGAm-*A AFPTCTGAA-CCTTTTTATTCTAG

16U

5'.•GATCT-GACTT-GG ATTTTCTGAA-CCTmTAMNiTTTTlTCTAG

5’.

GATCTAAAAGACTT-GG A’ATTTTCTGAA-CCTTTTTWm

CTAG

33L

j;.: J

B TOPI Q

GATCTAAAAGACTT ATTTTCTGAA-CCTTTTTAARAATmTTCTAG

5’.

5

GG

-

ATCTTTTTTA ATTTTCTGRA-CCTTTTTAAAA’

GATCTAAAAGACTT-G

C

137

induced DNA cleavage (Jaxel et al., 1991). The effects of camptothecin and other top1 inhibitors were investigated in these reactions since there has been some controversy regarding the effect of camptothecin on the DNA cleavage step of top1 (Svejstrup et al., 1991; Kjeldsen et al., 1992). Our data demonstrate that: (1) camptothecin generally blocks the DNA religation step of topl, while inhibition of the top1 cleavage step probably depends on the local DNA sequence; (2) camptothecin increases the formation of top1 recombination intermediates (‘suicide products’) such as those mimicking the replication intermediate that has been postulated to occur in the presence of camptothecin; (3) top1 can generate an intramolecular recombination product with only a single base pairing 3’ to the cleavage site.

Rl

2. Materials and methods 5’m


R2

Fig. 2. (A) Oligonucleotide substrates used in the present study. The underlined sequence in oligo 32U corresponds to the hexadecameric Tetrahymena rDNA sequence containing a strong top1 cleavage site. The base at the 5’ terminus of the potential cleavage site (position + 1) was mutated from A to G in order to enhance camptothecin-induced DNA cleavage (Jaxel et al., 1991). The top1 cleavage is indicated as a space and the bases flanking this site are numbered - 1 and + 1, respectively. Oligonucleotides are labeled according to the length of the labeled strand (U: upper, L: lower). (B) Top1 reaction products. The C and R oligonucleotides represent the cleavage and religation products, respectively, as discussed in the text.

The present study is a further investigation of the effects of camptothecin on DNA cleavage and religation activities of top1 using oligonucleotides. The oligonucleotides used (Fig. 2A) were designed from the Tetrahymena hexadecameric sequence containing a previously identified strong top1 cleavage site (Bonven et al., 1985). The adenine at position (+ 1) of the Tetrahymena hexadecameric sequence was replaced by a guanine in order to enhance camptothecin-

Materials, enzymes and drugs

Synthetic oligonucleotides were purchased from The Midland Certified Reagent Company (Midland, TX) or synthesized using a 392 DNA/RNA synthesizer CABi, Applied Biosysterns, Foster City, CA). Phage T4 polynucleotide kinase and terminal deoxynucleotidyl transferase were purchased from Gibco BRL (Grand Island, NY). y-3* P-ATP and cr-32P-cordycepin S-triphosphate were purchased from New England Nuclear (Boston, MA). Polyacrylamide was purchased from Bio-Rad, Inc. (Richmond, CA). Calf thymus type I DNA topoisomerase (top11 was purchased from Gibco BRL. Camptothecin and derivatives were provided by Drs. Wani and Wall (Research Triangle Institute, Research Triangle Park, NC), or obtained from the Drug Synthesis and Chemistry Branch, DTP, DCT, NCI, Bethesda, MD. Saintopin was a gift from Dr. Hirofumi Nakano (Tokyo Research Laboratories, Japan), and SN-38 a gift from Dr. Kiyoshi Terada (Yakult Honsha, Co. Ltd, Japan). The drugs were dissolved in dimethyl sulfoxide at 10 mM immediately before use and further dilutions were made in water.

138

Y Pommier et al. /Mutation

Research 337 (1995) 135-145

Radiolabeling and annealing of oligonucleotides 5’ and 3’ end labeling of single-stranded

Topoisomerase reactions

oligonucleotides was performed using T4 polynucleotide kinase with Y-~*P-ATP, and terminal deoxynucleotidyl transferase with 32P-labeled cordycepin, respectively. The latter procedure adds a labeled cordycepin molecule to the 3 terminus of the oligonucleotide. The kinase reaction took place at 37°C for 30 min and was stopped by a lo-min incubation at 70°C. Annealing with an excess of the complementary strand was performed in 20 ~1 of annealing buffer (10 mM Tris-HCl pH 7.8, 100 mM NaCl, 1 mM EDTA). The reaction mixture was heated to 95°C for 5 min and slowly chilled to 20°C. It was subsequently centrifuged through a Quick Spin G-25 Sephadex column to remove excess singlestranded oligonucleotides and stored at 4°C.

Reactions were performed in 10 ~1 reaction buffer (0.01 M Tris-HCl pH 7.5, 150 mM KCl, 5 mM MgCl,, 0.1 mM EDTA, 15 ,ug/ml bovine serum albumin). Substrate oligonucleotides (approximately 50 fmol/reaction) were incubated with 10 units of top1 for 15 min at room temperature (unless otherwise indicated). In the intermolecular strand invasion experiments (Fig. 51, a two-fold excess of acceptor oligonucleotides was added and reactions were performed for an additional 30 min at room temperature. Reactions were stopped as indicated by adding either sodium dodecyl sulfate (SDS) or NaCl (0.5% and 0.5 M, respectively, as final concentrations). Proteinase K (ICN BiochemicaIs, Cleveland, OH) (0.5 mg/ml final concentration) was then added to the reaction mixtures and proteolysis was carried out for

Top I + CPT

Top I

DNA

alone

III

ABCDEF

ABCDEFGHIJKLMNO

33+ 4-33

32--

r

I1 16U

32U 1

e

*

&U

33u

*

33 L

Fig. 3. Topl-mediated DNA cleavage and religation using duplex and partially single-stranded oligonucleotides. The oligonucleotides used are schematized under the gel pictures according to Fig. 2 with the ‘,P label shown as *. R and C indicate the religation and cleavage products, respectively. Arrows and numbers correspond to the migration position and size of the DNA fragments. (A) ‘Lanes A and D: DNA alone; lanes B and E: + topl; and lanes C and F: +topl + 100 FM camptothecin. R indicates the religation products, C the cleavage products, and the numbers correspond to the migration position and size of the DNA substrates. (B) Lanes A and M: DNA alone; lanes B and N: + topl; lanes C and 0: + top1 + 100 PM camptothecin (CPT). Each set of three lanes in the middle corresponds to different stops: SDS only: lanes D, G, and J; SDS + proteinase K: lanes E, H, and K; and 0.5 M NaCl without proteinase K: lanes F, I, and L. The top of the gel has been cut but showed bands near the wells in lanes D, F, G, and I corresponding to topl-linked DNA.

Y. Pommier et al. /Mutation

an additional 60 min at 50°C. Proteolysis was halted by the addition of 36 ~1 2.5 X loading buffer (98% formamide, 0.01 M EDTA, 1 mg/ml xylene cyan01 and 1 mg/ml bromophenol blue). Gel electrophoresis, imaging and quantitation Twenty-five percent sequencing gels (25% acrylamide, 0.1% Bisacrylagel (National Diagnostics, Manville, NJ), 7 M urea) in TBE buffer (89 mM Tris-HCl pH 8, 89 mM boric acid, 2 mM EDTA) were run at 40 V/cm, 50°C for 2-3 h. At the end of electrophoresis, gels were transferred to Whatman 3 MM paper sheets and dried. Imaging and quantification were performed using a Phosphorimager (Molecular Dynamics, Sunnyvale, CA).

3. Results and discussion Topl-mediated cleotides

DNA cleavage in duplex oligonu-

Using the 32-mer oligonucleotide labeled at the 5’ terminus of the scissile strand (oligo 32U, Fig. 2A), we found that top1 generated a cluster of bands (labeled C for cleavage) (Fig. 3A, lane C) migrating slightly slower than a 16-mer (lane D). Camptothecin markedly enhanced this effect (compare lanes C and B). Since these DNA bands were retained near the gel wells in the absence of proteinase K treatment (not shown), they probably represent topl-linked DNA fragments. In order to analyze unambiguously the top1 cleavage products in this 32-mer oligo, the same strand that was 5’ end labeled in the above experiments was labeled by incorporating 32P-cordycepin at the 3’ terminus of the same strand (see Fig. 2A, oligo 33U). Under these conditions, the topl-generated DNA fragments are not linked to top1 and bear 5’-OH termini. Fig. 3B (lanes N and 0) shows that cleavage of the 33U oligo generated a prominent 19-mer fragment indicating that a single cleavage site was observed at the same position as in the Tetrahymena hexadecameric sequence (Bonven et al., 1985). The cluster of topl-induced DNA bands observed using 5’ labeling (oligo 32U, Fig. 3A, lane C) is therefore probably due to the incomplete digestion of the

Research 337 (1995) 135-145

139

top1 covalently linked to the 3’-DNA terminus after proteinase K treatment. Topl-mediated DNA cleavage in partially singlestranded oligonucleotides

Using oligo 16U (Fig. 2A), top1 in the presence and absence of camptothecin produced the same DNA bands as with the duplex 32U oligo (Fig. 3A, compare lanes F and E to lane 0. Top1 cleavage is therefore unaltered by the partially single-stranded character of the substrate, and occurs in the double-stranded portion of the molecule with only two base pairs 3’ to the site of cleavage (see Fig. 2B, C oligonucleotide). These results confirm (Svejstrup et al., 1991; Shuman, 1992a) that a partially single-stranded substrate is efficiently cleaved by eukaryotic top 1. Differences were observed between the duplex (32U, 33U) and the partially single-stranded oligonucleotides (16U). First, in the absence of camptothecin, top1 cleavage was more intense in the 16U oligo than in the 32U oligo (Fig. 3A, compare lanes E and B). Secondly, addition of 0.5 M NaCl reversed cleavage completely in the duplex oligonucleotide but only marginally in the 16U oligonucleotide (not shown). And third, the enhancing effect of camptothecin on topl-induced DNA cleavage appeared less in the 16U than in the 32U oligo (Fig. 3). Taken together these data are consistent with the formation of irreversible top1 cleavable complexes (‘suicide reactions’) in the oligonucleotide (16I.I) missing most of the scissile strand on the 3’ side of the cleavage site (Svejstrup et al., 1991; Shuman, 1992a). Dissociation of the dinucleotide GG 3’ to the cleavage site (see C in Fig. 2) prevents spontaneous religation, which may account for the higher level of cleavage observed in the absence of camptothecin in the partially single-stranded than in the duplex oligonucleotide. The enhancement of top1 cleavage by camptothecin in the suicide oligonucleotide where religation is minimal does not support the view that camptothecin generally suppresses the cleavage step of top1 reactions (Kjeldsen et al., 1992). Rather, cleavage enhancement is consistent with the view that camptothecin primarily blocks the religation step of top1 reactions.

Y Pommier et al. /Mutation Research 337 (1995) 135-145

140

Topl-mediated DNA recombination in the partially single-stranded oligonucleotide

only religates S-OH DNA termini, the 33L oligonucleotide was differentially phosphorylated at its 5’ termini in order to block religation of one or both strands at a time. Fig. 4 shows that the S-hydroxyl terminus from the lower strand but not from the upper strand was used as acceptor. Since both termini have the same base sequence (see Fig. 2A), the lack of religation of the upper strand might be due to its short single-stranded overhang (4 nucleotides vs. 20 nucleotides for the lower strand) (see reaction product R2 in Fig. 2B). It is also possible that religation is intramolecular between the 5’-G of the lower strand and the - 1 T of the upper strand of oligos 33L or 16U (see oligo Rl in Fig. 2B). Since only the 5’-G of the acceptor was complementary to the single-stranded part of the donor, our results indicate that topl-mediated DNA strand transfer reactions can take place with only one base complementarity between donor and acceptors. Formation of a DNA hairpin even in the absence of base complementarity at the + 1 position has been observed with the vaccinia top1 @human, 1992a). Christiansen et al. (1993) found in their oligonucleotide that eukaryotic top1 required two complementary nucleotides for religation of single-stranded acceptor DNA to a partially single-stranded donor. They also showed that two complementary nucleotides were not required in the case where both the DNA donor (covalently linked to top11

Another slowly migrating DNA product (labeled R in Fig. 3 and other figures) was generated by top1 with the partially single-stranded (16U) oligonucleotide. This R product was best detectable in the absence of camptothecin (Fig. 3A, lane E). Its electrophoretic migration coincided with a 46-mer and was similar when reactions were stopped with or without SDS or proteinase K (Fig. 3B, compare lanes D and F to lane E; Fig. 4, lane B), indicating that this DNA was not covalently linked to topl. 3’ labeling of the complementary strand with cordycepin (oligo 33L, Fig. 2A) generated a 47-mer major band (Fig. 3B, lanes B and C). Therefore, this ‘R’ product is most likely due to the religation of the cleaved upper strand with the 5’-hydroxyl of the lower strand (see Fig. 2B, oligos Rl and R2). This result is consistent with previous observations with both the eukaryotic and vaccinia virus enzymes (Halligan et al., 1982; Svejstrup et al., 1991; Shuman, 1992a) showing that top1 can catalyze a strand transfer reaction between the nucleotide covalently linked to the enzyme (donor) and the S-OH of an exogenous single-stranded DNA acceptor. Additional experiments were performed to confirm the selective use of the lower strand as acceptor in the topl-mediated DNA strand transfer reaction (Fig. 4). Because top1 generates and

,D,E,

,A,B,C, 1

,

I

I,

F I

I

,,G,H,I,,J,K,L, m.

.

.

-.

-

Fig. 4. Preferential religation of the long single-stranded end of the partially single-stranded oligonucleotide. The 33L oligonucleotide (see Fig. 2) was phosphorylated at the indicated 5’ termini as schematized above the gel picture. 32P-labeling is shown as * and phosphotylation of the 5’ DNA termini as a P in a circle. Such phosphorylation prevents the topl-mediated religation of this terminus. Sets of lanes A, D, G, and J: DNA alone; sets of lanes B, E, H, and K: + topl; sets of lanes C, F, I, and L: + top1 + 100 PM camptothecin. Each set of lanes contains three reactions stopped (from left to right) with SDS, SDS + proteinase K, and 0.5 M NaCI. R and 33 indicate the migration position of the religation and DNA substrate, respectively.

Y. Pommier et al. /Mutation

and acceptors were blunt ended, and postulated a ‘bipartite’ interaction of top1 with both the acceptor and the donor that would favor the juxtaposition of the two ends. A bipartite interaction in our oligonucleotide is not obvious since the single-stranded acceptor does not form a duplex structure with the single-stranded tail of the donor (Fig. 2B, oligo Rl). Nevertheless, our observation implies that top1 can act as an efficient strand transferase and mediate facile DNA recombinations when a single-stranded DNA fragment bearing a S-OH invades a top1 recombination intermediate (e.g. lacking its S-DNA acceptor) (Halligan et al., 1982) (see Fig. 10 Differential religation of singlestranded acceptor oligonucleotides

and

141

Research 337 (1995) 135-145

double-

Using the partially single-stranded (16U) oligonucleotide, we tested the religation of a duplex acceptor that would mimic the camptothecin-induced DNA damage at replication forks (Fig. 1A). Intramolecular religation was prevented by phosphorylating the 5’ terminus of the lower strand of the oligonucleotide (Fig. 4, lanes G-I; Fig. 5, oligo A). The single-stranded

acceptor complementary to the single-stranded tail of the 16U oligonucleotide was religated, as indicated by the appearance of a band migrating as a 32-mer oligonucleotide (Fig. 5, oligo and lane Sl @human, 1992b; Christiansen et al., 1993). By contrast, no detectable religation was detected in the presence of the duplex acceptor (Fig. 5, oligo and lane D). Religation of such duplex acceptors has been shown to require that the single-stranded tail of the donor be absent or less than two nucleotides @human, 1992b; Christiansen and Westergaard, 1994). Together, the results obtained in oligonucleotides demonstrate that top1 recombination intermediates at a single-strand break site inefficiently religate a duplex DNA acceptor. They are consistent with the observation of persistent topl-induced DNA doublestrand breaks in replicating DNA of cehs treated with camptothecin (Ryan et al., 1991) and SV40 DNA (Tsao et al., 1993) (see Fig. 1A). Effects of camptothecin and other top1 inhibitors

As indicated above, camptothecin had two effects in the partially single-stranded oligonucleotide: cleavage enhancement and strand transA

16

B

S

D

5’-•GATCTAAAAGACTT-GG ATTTTCTGAA-CCTTTTTAAAAATTTTTTCTAG@

Q .$??j$

C

GG

4 32

5’-‘GATCTAUAGACTT ATTTTCTGAA-CCTTTTTAAAAATTTTTTCTAG

32 5’-+GATCTAAAAGACTT+GAAAAATTlYTAMAM) ATTTTCTGAA-CCTTTTTAAAAATTTTTTCTAG@

CCTTTTTAAMATTTTTTCTAG GAAAAATTTTTAMAM 5’-*GATCTAAAAGACTT ATTTTCTGAA-CCTTTTTAAAAATTTTTTCTAG

-C I 16

Fig. 5. Lack of religation of a duplex DNA acceptor. (A) The 16U oligonucleotide (see Fig. 2) labeled at the 5’ terminus of the upper strand was phosphorylated at the 5’ terminus of the lower strand (P in circle) to avoid intramolecular religation. (B) The 16U oligonucleotide was reacted for 15 min at 20°C to form the topl-linked DNA cleavage product (labeled 0. (S) Addition of a single-stranded donor oligonucleotides (boxed sequences in the left panel) for 30 min at 20°C yielded a 32-mer religation product. (D) Addition of the double-stranded oligonucleotide failed to produce detectable rehgation.

142

Y. Pommier et al. /Mutation Research 337 (1995) 135-145

fer inhibition (Fig. 3 and Fig. 4). The camptothecin concentration-response curves of top1 cleavage enhancement and inhibition of strand transfer are shown in Fig. 6. Camptothecin concentrations around 1 PM inhibited religation by approximately 50%. Concurrently, top1 cleavage was enhanced as camptothecin concentration increased indicating that camptothecin can increase top1 recombinogenic intermediates. The camptothecin concentrations which were effective in the presently described top1 reactions are in the same range as those reported to stimulate top1 cleavage and protein-linked DNA breaks in cells (Covey et al., 1989; Tanizawa et al., 1994). The finding of concomitant inhibition of strand transfer and enhancement of cleavage is consistent with the view that the major effect of camptothecin on top1 is to stabilize the topl-DNA cleavable complexes by inhibiting their resealing (Hsiang et al., 1985; Svejstrup et al., 1991; Pommier and Tanizawa, 1993). Time course experiments using 10 PM camptothecin showed that the inhibitory effect of camptothecin on toplDNA strand transferase activity (religation) could be overcome by increasing the reaction time (data not shown). Therefore, inhibition of top1 DNA strand transfer activity by camptothecin appears reversible. Our data show that the previously reported inhibition of suicide top1 cleavage by camptothecin (Kjeldsen et al., 1992) cannot be generalized to all top1 cleavage sites since such an effect could not be detected in our oligonucleotide. The discrepancy may be due to differences in oligonucleotide sequences as Kjeldsen and coworkers used a special suicide substrate of what they called ‘the class b’ recognition sequences, cleavage of which is strongly stimulated by camptothecin. In any case, inhibition of top1 religation is generally observed in all oligonucleotides studied so far and in longer DNA fragments with various camptothecin derivatives (Porter and Champoux, 1989; Tanizawa et al., 1995). Therefore, it is safe to conclude that the major common effect of camptothecin is to inhibit top1 religation. Other camptothecin derivatives, lO,ll-methylenedioxy-20-S-camptothecin, 9-aminocamptothe-

pM CPT I A

A

B 0 0.1 0.3 I

3

1

IO 30 loo

R-

B

35 . g

30: 25:

i? 20

P

E.

15

0

2

4

6

8

10

pM CPT Fig. 6.

followed by proteinase product, C the cleavage position of the 16U oligonucleotide substrate. (B) Quantification cleavage and religation presence of increasing camptothecin (CPT) concentrations.

mediated DNA strand DNA cleavage

transfer

enhanced

k: Pommier et al. /Mutation

1995). Saintopin, a dual inhibitor of topoisomerases I and II (Yamashita et al., 1991; Leteurtre et al., 1994) was also active (58% inhibition of relaxation and 1.2-fold enhancement of cleavage). These results indicate that the cleavage and religation assays using partially singlestranded oligonucleotides are sensitive to top1 inhibitors.

Research 337 (1995) 135-145

143

1992a,b; Christiansen et al., 1993). Such DNA structures could serve as substrates for intermolecular strand transfer with another 5’-hydroxyl DNA terminus (Fig. 1C). Inhibition of religation by camptothecin may only be limited since the drug effect is reversible by lowering camptothecin concentration and with time. The potential mutagenicity of camptothecin should be kept in mind when using this drug and its derivatives in the clinic.

4. Conclusions The present study shows the effect of camptothecin on the induction of DNA recombinations by eukaryotic top1 in simple oligonucleotides mimicking top1 recombinogenic intermediates at DNA single-strand break sites. The DNA strand transferase activity of top1 is reIatively efficient since it can take place in the absence of significant complementarity between the invading strand (acceptor) and the singlestranded segment of the topl-linked DNA (donor). Two main factors appear required for topl-mediated DNA recombinations: (1) an irreversibly cleaved topl-DNA complex (recombination intermediate), and (21 a single-stranded DNA fragment bearing a S-OH (see Fig. 10 This strand transferase activity of top1 may be physiologically important. For instance, it has been invoked for the genomic integration of SV40 (Bullock et al., 1985) and hepadnaviruses (Wang and Rogler, 1991). The present study may also contribute to a better understanding of the mutagenic effects of camptothecin (Degrassi et al., 1989; Backer et al., 1990; Pommier and Bertrand, 1993; Anderson and Berger, 1994). We observed that camptothecin increases the formation of recombinogenie top1 cleavable complexes by inhibiting the legitimate religation step of topl. Such recombination intermediates have been postulated after collisions of DNA replication forks with topl-linked DNA breaks (Fig. 1A) (Holm et al., 1989; Liu, 1989; Pommier and Tanizawa, 19931, and when the top1 cleavable complex is located within six nucleotides from a nick on the same DNA strand (Fig. 1B) (Svejstrup et al., 1991; Shuman,

Acknowledgements

The authors wish to thank Dr. Kurt W. Kohn, Chief, Laboratory of Molecular Pharmacology, NC1 for his suggestions and support during the course of this work. Thanks also to Drs. Abhijit Mazumder for stimulating discussions. We also wish to acknowledge Drs. Monroe E. Wall and Mans&h C. Wani, Chemistry and Life Sciences, Research Triangle Institute, Research Triangle Park, NC for the generous gift of camptothecin derivatives, Dr. Hirofumi Nakano, Kyowa Hakko Co. Ltd, Tokyo Research Laboratories, Tokyo, Japan for the gift of saintopin, and Dr. Kiyoshi Terada, Yakult Honsha Co. Ltd, Tokyo, Japan for the generous gift of SN-38.

References Anderson, R.D. and N.A. Berger (1994) Mutagenicity and carcinogenicity of topoisomerase-interactive drugs, Mutation Res., 309, 109-142. Backer, L.C., J.W. Allen, K. Harrington-Brock, J.A. Campbell, D.M. DeMarini, C.L. Doerr, D.R. Howard, A.D. Kligerman and M.M. Moore (1990) Genotoxicity of inhibitors of DNA topoisomerase I (camptothecin) and II (m-AMSA) in vivo and in vitro, Mutagenesis, 5, 541-547. Been, M.D. and J.J. Champoux (1984) Breakage of singlestranded DNA by eukaryotic type 1 topoisomerase occurs only at regions with the potential for base-pairing, J. Mol. Biol., 180, 515-531. Been, M.D., R.R. Burgess and J.J. Champoux (19841 Nucleotide sequence preference at rat liver and wheat germ type 1 DNA topoisomerase breakage sites in duplex SV40 DNA, Nucleic Acids Res., 12, 3097-3114. Bonven, B.J., E. Gocke and 0. Westergaard (198.5) A high affinity topoisomerase I binding sequence is clustered at

144

Y. Pommier et al. /Mutation Research 337 (1995) 135-145

DNAase I hypersensitive sites in Tetrahymena R-chromatin, Cell, 41, 541-551. Bullock, P., J.J. Champoux and M. Botchan (1985) Association of crossover points with topoisomerase I cleavage sites: a model for nonhomologous recombination, Science, 230,954-958. Champoux, J. (1990) Mechanistic aspects of type-l topoisomerases, in: J.C. Wang and N.R. Cozarelli (Eds.), DNA Topology and its Biological Effects, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, pp. 217-242. Champoux, J.J. and P.A. Bullock (1988) Possible role for the eucaryotic type I topoisomerase in illegitimate recombination, in: R. Kucherlapati and G.R. Smith (Eds.), Genetic Recombination, American Society for Microbiology, Washington, DC, pp. 655-666. Christiansen, K. and 0. Westergaard (1994) Characterization of intra- and intermolecular DNA ligation mediated by eukaryotic topoisomerase I, J. Biol. Chem., 269, 721-729. Christiansen, K., B.D. Svejstrup, A.H. Andersen and 0. Westergaard (1993) Eukaryotic topoisomerase I-mediated cleavage requires bipartite DNA interaction, J. Biol. Chem., 268, 9690-9701. Covey, J.M., C. Jaxel, K.W. Kohn and Y. Pommier (1989) Protein-linked DNA strand breaks induced in mammalian cells by camptothecin, an inhibitor of topoisomerase I, Cancer Res., 49. 5016-5022. Degrassi, F., R. De Salvia, C. Tanzarella and F. Palitti (1989) Induction of chromosomal aberrations and SCE by camptothecin, an inhibitor of mammalian topoisomerase I, Mutation Res., 211, 125-130. Halligan, B.D., J.L. Davis, K.A. Edwards and L.F. Liu (1982) Intra- and intermolecular strand transfer by HeLa DNA topoisomerase I, I. Biol. Chem., 257, 3995-4000. Holm, C., J.M. Covey, D. Kerrigan and Y. Pommier (1989) Differential requirement of DNA replication for the cytotoxicity of DNA topoisomerase I and II inhibitors in Chinese hamster DCSF cells, Cancer Res., 49, 6365-6368. Hsiang, Y.H., R. Hertzberg, S. Hecht and L.F. Liu (198.5) Camptothecin induces protein-linked DNA breaks via mammalian DNA topoisomerase I, J. Biol. Chem., 260, 14873-14878. Hsiang, Y.-H., M.G. Lihou and L.F. Liu (1989) Arrest of DNA replication by drug-stabilized topoisomerase I-DNA cleavable complexes as a mechanism of cell killing by camptothecin, Cancer Res., 49, 5077-5082. Hsieh, T.-S. (1992) DNA topoisomerases, Curr. Biol. 4, 396400. Jaxel, C., K.W. Kohn and Y. Pommier (1988) Topoisomerase I interaction with SV40 DNA in the presence and absence of camptothecin, Nucleic Acids Res., 16, 11157-11170. Jaxel, C., G. Capranico, D. Kerrigan, K.W. Kohn and Y. Pommier (1991) Effect of local DNA sequence on topoisomerase I cleavage in the presence or absence of camptothecin, .I. Biol. Chem., 266, 20418-20423. Kjeldsen, E., S. Mollerup, B. Thomsen, B.J. Bonven, L. Bolund and 0. Westergaard (1988) Sequence-dependent ef-

feet of camptothecin on human topoisomerase I DNA cleavage, J. Mol. Biol., 202, 333-342. Kjeldsen, E., J.Q. Svejstrup, I. Gromova, J. Alsner and 0. Westergaard (1992) Camptothecin inhibits both the cleavage and religation reactions of eukaryotic DNA topoisomerase I, J. Mol. Biol., 228, 1025-1030. Leteurtre, F., A. Fujimori, A. Tanizawa, A. Chhabra, A. Mazumder, G. Kohlhagen, H. Nakano and Y. Pommier (1994) Saintopin, a dual inhibitor of DNA topoisomerases 1 and II, as a probe for drug-enzyme interactions, J. Biol. Chem., 269, 28702-28707. Liu, L.F. (1989) DNA topoisomerase poisons as antitumor drugs, Annu. Rev. Biochem., 58, 351-375. Pommier, Y. and R. Bertrand (1993) The mechanisms of formation of chromosomal aberrations: role of eukaryotic DNA topoisomerases, in: I.R. Kirsch (Eds.), The Causes and Consequences of Chromosomal Aberrations, CRC Press, Boca Raton, FL, pp. 277-309. Pommier, Y. and A. Tanizawa (1993) Mammalian DNA topoisomerase I and its inhibitors, in: J. Hickman and T. Tritton (Eds.1, Cancer Chemotherapy, Blackwell Scientific, Oxford, pp. 214-250. Porter, S.E. and J.J. Champoux (1989) The basis for camptothecin enhancement of DNA breakage by eukaryotic topoisomerase I, Nucleic Acids Res., 17, 8521-8532. Potmesil, M. (1994) Camptothecins: from bench research to hospital wards, Cancer Res., 54, 1431-1439. Ryan, A.J., S. Squires, H.L. Strutt and R.T. Johnson (1991) Camptothecin cytotoxicity in mammalian cells is associated with the induction of persistent double strand breaks in replicating DNA, Nucleic Acids Res., 19, 3295-3300. Shuman, S. (1992a) DNA strand transfer reactions catalyzed by vaccinia topoisomerase I, J. Biol. Chem., 267, 86208627. Shuman, S. (1992b) Two classes of DNA end-joining reactions catalyzed by vaccinia topoisomerase I, J. Biol. Chem., 267, 16755-16758. Slichenmyer, W.J., E.K. Rowinsky, R.C. Donehower and S.H. Kaufmann (1993) The current status of camptothecin analogs as anticancer agents. J. Natl. Cancer Inst., 85. 271-287. Snapka, R.M. and P.A. Permana (1993) SV40 DNA replication intermediates: analysis of drugs which target mammalian DNA replication, BioEssays, 15, 121-127. Svejstrup, J.Q., K. Christiansen, 1.1. Gromova, A.H. Andersen and 0. Westergaard (1991) New technique for uncoupling the cleavage and religation reactions of eukaryotic topoisomerase I. The mode of action of camptothecin at a specific recognition site, J. Mol. Biol., 222, 669-678. Tanizawa, A., A. Fujimori, Y. Fujimori and Y. Pommier (1994) Comparison of topoisomerase I inhibition, DNA damage, and cytotoxicity of camptothecin derivatives presently in clinical trials, J. Natl. Cancer Inst., 86, 836842. Tanizawa, A., K.W. Kohn, G. Kohlhagen, F. Leteurtre and Y. Pommier (1995) Differential stabilization of eukaryotic

Y Pommier et al. /Mutation Research 337 (1995) 135-145

DNA topoisomerase I cleavable complexes by camptothecin derivatives, Biochemistry, 43, 7200-7206. Tsao, Y.P., A. Russo, G. Nyamusa, R. Silber and L.F. Liu (1993) Interaction between replication forks and topoisomerase I-DNA cleavable complexes: studies in a cell-free SV40 DNA replication system, Cancer Res., 53,5908-5914. Wang, H.-P. and C.E. Rogler (1991) Topoisomerase I-media-

145

ted integration of hepadnavirus DNA in vitro. J. Viral., 65, 2381-2392. Yamashita, Y., S.-Z. Kawada, N. Fujii and H. Nakano (1991) Induction of mammalian topoisomerase I and II mediated DNA cleavage by saintopin, a new antitumor agent from fungus, Biochemistry, 30. 58385845.