Journal of Cell Science 110, 1345-1350 (1997) 1345 Printed in Great Britain © The Company of Biologists Limited 1997 JCS9577 COMMENTARY When helicase and topoisomerase meet! Michel Duguet Laboratoire d’Enzymologie des Acides Nucléiques, Institut de Génétique et Microbiologie, URA 2225 CNRS, Université Paris-Sud, 91405 Orsay Cedex, France (e-mail: [email protected]) SUMMARY Several examples of direct interactions between helicases finding of an interaction between type II topoisomerase and and topoisomerases have recently been described. The data the helicase Sgs1 in yeast. This interaction possibly allows suggest a possible cooperation between these enzymes in the faithful segregation of newly replicated chromosomes in major DNA events such as the progression of a replication eukaryotic cells. A fourth example is the interaction between fork, segregation of newly replicated chromosomes, disrup- the same helicase Sgs1 and topoisomerase III in yeast, that tion of nucleosomal structure, DNA supercoiling, and finally may control recombination level and genetic stability of recombination, repair, and genomic stability. A first example repetitive sequences. Recently, in humans, mutations in is the finding of a strong interaction between T antigen and genes similar to Sgs1 have been found to be responsible for topoisomerase I in mammalian cells, that may trigger Bloom’s and Werner’s syndromes. The cooperation between unwinding of the parental DNA strands at the replication helicases and topoisomerases is likely to be extended to many forks of Simian Virus 40. A second example is the reverse aspects of DNA mechanisms including chromatin condensa- gyrase from thermophilic prokaryotes, composed of a tion/decondensation. putative helicase domain, and a topoisomerase domain in the same polypeptide. This enzyme may be required to maintain Key words: Replication fork, Chromosome segregation, genomic stability at high temperature. A third example is the Nucleosome, DNA supercoiling, Recombination, Genetic stability INTRODUCTION Several recent examples of interactions of helicases with topoisomerases suggest a cooperation between these two Early in evolution, the choice was made to have the plectone- classes of enzymes in many aspects of DNA mechanisms (Fig. mic DNA double helix as the carrier of genetic information. 1). These include progression of the DNA replication fork, seg- From this initial choice, it was necessary to solve the topo- regation of newly replicated chromosomes, disruption of logical problem of separating the two DNA strands in order nucleosomal structure (especially during transcription), DNA to allow the proper transmission of the genetic information. supercoiling, and finally DNA recombination, repair, and Two steps are required for this process: (1) disruption of genome stability. hydrogen bonds between the two strands, performed by spe- cialized enzymes called DNA helicases (reviewed by Lohman, 1993); and (2) elimination of all the topological links between PROGRESSION OF THE REPLICATION FORK the two strands, performed by DNA topoisomerases (reviewed by Wang, 1996). These considerations led to the suggestion In the late seventies, it was suggested that the topological that DNA helicases and topoisomerases are both required to problems arising at a replication fork could in principle be solved provide the swivel mechanism for DNA replication, postulated by two different mechanisms (Champoux and Been, 1980; more than three decades ago by Cairns (1963). Forterre et al., 1980). The first consists of the removal of the Surprisingly, various models for the replisome, a compact positive supercoils arising in front of the replication fork, as the replication complex comprising DNA polymerase/primase and parental strands rotate. In eukaryotic cells, this may be achieved helicases, did not generally integrate a topoisomerase indifferently by topoisomerase I or II (Kim and Wang, 1989; (Kornberg, 1978). The first indication of a cooperation between DiNardo et al., 1984), although the former enzyme appears the a helicase and a topoisomerase came from studies on phage more efficient because of its highly processive relaxing activity ΦX174 DNA replication. An interaction between the host (Roca, 1995). In prokaryotic cells, DNA gyrase (topoisomerase encoded rep DNA helicase and the phage gene A protein II), efficiently removes positive supercoils, and may function product (a sequence specific topoisomerase) allows for strand ahead of the fork (Wu et al., 1988), while topoisomerase I separation of the phage DNA replicative form (Scott and (protein ω) is probably less involved, as it is unable to bind pos- Kornberg, 1978; Duguet et al., 1978). itively supercoiled DNA (Kirkegaard and Wang, 1985). Fig. 2A 1346 M. Duguet depicts a hypothetical swivelase (see also Fig. 5) formed by the would explain why a defect in a helicase that does not interact association of a helicase and a topoisomerase. In this model, the with topoisomerase II does not result in chromosome unstability two enzymes are part of a large replication complex anchored in (i.e. XPD or ERCC2). a fixed structure (i.e. the nuclear matrix) and the replicating DNA Very recently, two human genes, BLM and WRN (respec- is translocated through the protein complex (Kornberg, 1988). tively 43% and 38% identical to Rec Q, and 44% and 34% to The helicase actively separates the two parental DNA strands SGS-1 over the 400 amino acid sequence of the helicase while the topoisomerase, working in front of the helicase, allows region) have been identified as responsible for Bloom’s and relaxation of positive supercoils in a highly processive manner. Werner’s syndromes (Ellis et al., 1995; Yu et al., 1996). Similar Indeed, a direct interaction between eukaryotic topoisomerase I to mutations in SGS-1, mutations in BLM and WRN result in and T antigen was recently described, that may trigger DNA chromosomal breakage, translocations, intra and interchromo- unwinding at the SV40 replication forks (Simmons et al., 1996). somal strand exchange. Furthermore, a reduction in 1 M NaCl The second mechanism to prevent positive supercoiling at the extractible topoisomerase II was found in Bloom’s cells treated replication fork allows newly replicated DNA duplexes to by BrdU (Heartlein et al., 1987). untangle (Champoux and Been, 1980; Forterre et al., 1980), Finally, it is noteworthy that SGS-1 (and perhaps the human instead of rotating the parental duplex. This may be achieved by genes) not only interact in vivo with topoisomerase II, but also a type II topoisomerase working behind the helicase and able to in a distinct pathway with topoisomerase III, an interaction that perform transient double strand breaks in the DNA. Such a is presumably required to reduce the level of recombination mechanism is likely to occur at the end of replication (see below). (see section on recombination and genome stability). SEGREGATION OF NEWLY REPLICATED DISRUPTION OF NUCLEOSOMAL STRUCTURE CHROMOSOMES Replication and transcription require that the DNA be accessible It has been proposed long ago that proper chromosome segre- to the enzymatic machineries involved in these processes. It was gation presents a major topological problem, both in eukaryotic suggested, at least for the initiation steps, that nucleosomal struc- and in prokaryotic cells, because of the intertwining of newly tures must be transiently disrupted, then reassembled on newly replicated DNA molecules (Sundin and Varshavsky, 1980). replicated DNA or after passage of a transcription complex Failure to solve this problem results in breakage of chromo- (Bonne-Andrea et al., 1990; Adams and Workman, 1993). somes, non-disjunction, and eventually cell death (Holm et al., Nucleosome disruption may occur via the positive super- 1989; Uemura et al., 1987; Spell and Holm, 1994). The essential coiling produced by a helicase tracking through the DNA role of type II topoisomerases (Topo II in eukaryotes, gyrase or duplex in the presence of a topoisomerase. This idea is based Topo IV in bacteria) in the untangling of chromosomes has been on the twin supercoiled model imagined by Liu and Wang well recognized (DiNardo et al., 1984; Adams et al., 1992). (1987) for transcription. In the hypothetical model in Fig. 2C, Fig. 2B shows schematically the problem arising when two the helicase locally separates the DNA strands in front of the forks meet at the end of replication (Champoux and Been, 1980; topoisomerase, producing positive supercoils ahead and Forterre et al., 1980; Wang, 1991; Holm, 1994). As pointed out negative supercoils behind (Liu and Wang, 1987; Koo et al., by Holm, when the tracking DNA helicases approach each other, 1991). Positive supercoiling is absorbed by nucleosome dis- any double helical turn in the parental duplex is converted into ruption, while negative supercoiling is removed by the topoi- one intertwining between the two daughter molecules. Topoiso- somerase, allowing the double helix to rewind. This implies merase II removes these intertwinings by its double-strand that the topoisomerase acting in this process may selectively breaking/passing/rejoining activity (Wang, 1996). Thus, a remove negative supercoils (like prokaryotic topo I), or that helicase working at a replication fork and a type II topoisomerase positive supercoils are not targeted for immediate removal. working in association with it, may cooperate to perform chro- Nucleosome disruption was recently demonstrated by mosome segregation. Until recently, this cooperation remained a Ramsperger