A human DNA helicase homologous to the DNA cross-link sensitivity protein Mus308.

Repair of DNA interstrand cross-links is a challenging problem for cells. Many human gene products influence sensitivity to DNA cross-linking agents, but the mechanisms of cross-link repair are unknown. In Drosophila melanogaster, the mus308 mutation leads to marked sensitivity to DNA cross-linking agents. The C-terminal portion of the Mus308 polypeptide encodes a DNA polymerase, whereas a putative DNA helicase is encoded by the N-terminal portion. As a step toward isolating proteins involved in DNA cross-link repair, we searched for mammalian genes similar to the DNA helicase portion of Mus308. Human and mouse homologs were isolated from cDNA expression libraries and designated HEL308. Human HEL308 is on chromosome 4q21 and encodes a polypeptide of 1101 amino acids. The protein was expressed in insect cells and purified. HEL308 is a single-stranded DNA-dependent ATPase and DNA helicase. Mutation of a highly conserved lysine to methionine in helicase domain I eliminated both activities. The protein readily displaces 20- to 40-mer duplex oligonucleotides. Displacement of longer substrates was less efficient but was stimulated by the single-stranded DNA-binding protein RPA. Activity was supported by ATP or dATP but not other nucleotide triphosphates. The enzyme translocates on DNA with 3' to 5' polarity and behaves as a multimer upon gel filtration.

DNA interstrand cross-linking agents such as nitrogen mustards, mitomycin C, and psoralen are widely used in cancer chemotherapy because of their high cytotoxicity to dividing cells (1). A single unrepaired interstrand cross-link (ICL) 1 can kill a bacterial or yeast cell, and about 40 unrepaired ICLs can kill a mammalian cell (2,3). Moreover, DNA ICLs induce mutations and chromosomal rearrangements. The most extensive studies of cross-link repair have been carried out in Escherichia coli, in which the major interstrand cross-link repair pathway is well characterized, both genetically and biochemically. ICL repair in eukaryotes is less well understood, and there may be several pathways (4). In E. coli and in Saccharomyces cerevisiae, the repair of ICLs depends on both nucleotide excision repair and homologous recombination (5)(6)(7). In mammals, mutant cell lines sensitive to cross-linking agents have been useful to identify proteins that might be involved in ICL repair. XPF and ERCC1 mutant cell lines, in addition to being defective in nucleotide excision repair, are particularly sensitive to crosslinking agents (8), suggesting that these proteins play a special role in ICL repair. Similarly, mutations in the XRCC2 and XRCC3 genes, encoding proteins with sequence homology to the human RAD51 protein, confer sensitivity to cross-linking agents (9). Fanconi anemia cell lines are also particularly susceptible to such agents. Thus far, eight Fanconi anemia complementation groups have been defined, and six genes have been mapped and cloned, FANCA, FANCC, FANCD2, FANCE, FANCF, and FANCG (1). Studying Fanconi anemia will likely be of great value in understanding human ICL repair mechanisms, but the function of the FANC proteins is still unclear.
In Drosophila melanogaster, mutations in the mus308 gene lead to marked sensitivity to cross-linking agents. Experiments suggested that some incision event takes place in mus308 mutants, but full repair does not take place (10). The C-terminal portion of the Mus308 protein encodes a DNA polymerase, whereas the N-terminal portion encodes the seven characteristic motifs found in DNA and RNA helicases (11). Sharief et al. (12) cloned a human cDNA for POLQ, encoding a polypeptide homologous to the Mus308 polymerase domain but with no corresponding helicase region. A longer cDNA sequence for human POLQ, deposited with the NCBI data base by Abbas and Linn (NCBI accession number NM_006596, predicts a presumably full-length protein with the polymerase domain in the C-terminal portion and a helicase domain at the N terminus, similar to D. melanogaster Mus308. In many DNA repair pathways, the function of DNA helicases is essential. In particular, UvrD helicase is needed to repair ICLs in E. coli (13). Moreover, defects in DNA helicases are the causes of several human diseases. BLM and WRN, the products of the Bloom and Werner syndrome genes, are members of the RecQ family of DNA helicases (14,15). Although their most critical roles in cells are not precisely known, they participate in pathways of DNA damage tolerance. Another member of the RecQ family of helicases, RECQ4, has been implicated in a subset of cases of Rothmund-Thompson syndrome (16). XPD and XPB helicases, two of the subunits of TFIIH transcription/repair factor, are involved in nucleotide excision repair (17), and mutations in their genes can lead to the disorders xeroderma pigmentosum and trichothiodystrophy.
With the aim of isolating new proteins implicated in DNA cross-link repair, we sought mammalian homologs of the putative helicase portion of D. melanogaster Mus308. We report * This work was supported by postdoctoral fellowships from the European Molecular Biology and from Telethon (to F. M.), by the Imperial Cancer Research Fund, and by the University of Pittsburgh Cancer Institute. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM  identification of new human and mouse genes and the biochemical activity of the human gene product.

EXPERIMENTAL PROCEDURES
Cloning of Human and Mouse HEL308 -Human and mouse HEL308 genes were cloned by 3Ј and 5Ј rapid amplification of cDNA ends using a CLONTECH SMART RACE cDNA Amplification kit. Primers were designed from human expressed sequence tags AA625285, H08004, and R24580 in order to clone the full human HEL308 coding region. cDNA was prepared from the testis cancer cell line 833K and the bladder cancer cell line MGH-U1 (18) according to the manufacturer's instructions (CLONTECH). Two fragments of 1767 bp (3Ј RACE) and 2167 bp (5Ј RACE) were combined to obtain the entire HEL308 coding region. The mutant HEL308 K365M gene was generated with the QuikChange site-directed mutagenesis kit (Stratagene); the single AAA to ATG (lysine to methionine) substitution was confirmed by sequencing. Primers were designed from mouse expressed sequence tag AA517170 to clone mouse Hel308 by RACE PCR from mouse leukemia L1210 cells RNA (19) and mouse total liver RNA (Ambion Inc.) using a CLONTECH SMART RACE cDNA Amplification kit.
Purification of Human HEL308 -The human HEL308 open reading frame was subcloned into plasmid pFastBac HTb, and the Bac-to-Bac baculovirus expression system (Invitrogen) was used to obtain recombinant baculovirus to infect Sf9 cells. A 500-ml spinner flask of Sf9 cells (1 ϫ 10 6 /ml) was infected with the His6-HEL308 baculovirus (multiplicity of infection of 2) for 48 h at 27°C. Cells were lysed in 20 ml of Buffer A (0.15 M Tris, pH 8.0, 0.15 M KCl, 1 mM phenylmethylsulfonyl fluoride, 10% glycerol, 0.5% Nonidet P-40, EDTA-free protease inhibitor mixture from Roche Diagnostics) for 30 min on ice. Insoluble material was removed by centrifugation (22,000 ϫ g for 15 min). KCl was added to raise the salt concentration to 0.3 M. The extract (12.2 mg/ml) was incubated overnight at 4°C with 5 ml of Ni 2ϩ -nitrilotriacetate superflow resin (Qiagen). The resin was washed with 5 column volumes of Buffer A plus 20 mM imidazole, and the protein was eluted with 20 ml (4 column volumes) of 100 mM imidazole. After reducing the salt concentration to 0.1 M with 10% (v/v) glycerol, the eluate was loaded onto a Mono Q column, HR 5/5 (AKTA design system, Amersham Biosciences, Inc.) equilibrated with Buffer C (20 mM Hepes-KOH, pH 8.0, 1 mM EDTA, 10% glycerol, 1 mM dithiothreitol, 0.01% Nonidet P-40) containing 0.1 M KCl. The column was washed with 1 column volume of equilibration buffer, and HEL308 protein was eluted with a 20-column volume linear gradient from 0.1 to 0.5 M KCl in Buffer C. HEL308 presence in the eluted fractions was detected by ATPase and helicase activities and by immunoblotting. Active fractions eluting from 0.275 to 0.425 M KCl were pooled, dialysed into Buffer C containing 0.1 M KCl, and loaded onto a Mono S column, HR 5/5 (AKTA design system, Amersham Biosciences, Inc.) equilibrated with Buffer C containing 0.1 M KCl. The column was washed with 1 column volume of equilibration buffer, and HEL308 protein was eluted with a 20-column volume linear gradient from 0.1 to 0.3 M in Buffer C. Active fractions eluting from 0.125 to 0.20 M KCl were pooled and dialysed into Buffer C containing 0.1 M KCl. Protein concentration was determined with the Coomassie Plus Protein Assay Reagent Kit (Pierce).
For analysis by gel filtration, 200 l of fraction number 28 from the Mono Q elution (ϳ0.3 M KCl) was loaded onto a Superose TM 6 HR 10/30 column (AKTA design system, Amersham Biosciences, Inc.) equilibrated in Buffer C without Nonidet P-40 and containing 0.5 M KCl. Protein was eluted with 1.5 column volumes of the same buffer at a flow rate of 0.5 ml/min, and 0.5-ml fractions were analyzed by immunoblotting, ATPase, and helicase assays. A calibration curve was prepared by measuring the elution volumes of molecular weight standards using a Gel Filtration Calibration Kit (Amersham Biosciences, Inc.).
ATPase Assay-Standard reaction mixtures (10 l) contained: 50 mM KCl, 20 mM Tris-HCl, pH 7.5, 4 mM MgCl 2 , 1 mM dithiothreitol, 50 g/ml bovine serum albumin, 0.1 mM cold ATP, 0.25 Ci [␥-32 P]ATP (specific activity Ͼ 5000 Ci/mmol), 1.4 nM HEL308 or 1.4 nM HEL308 K365M protein. Incubations were for 60 min at 30°C. Reactions were terminated by the addition of 5 l of 0.5 M EDTA. Released phosphate was separated from ATP by thin-layer chromatography on polyethyleneimine cellulose using 0.75 M KH 2 PO 4 as the running buffer. Hydrolysis was quantified with the use of a Fuji phosphorimaging device.
Accession Numbers-The GenBank TM accession number of D. melanogaster mus308 is L76559. The human POLQ NCBI accession number is NM_006596. The mus1 homolog from Caenorhabditis elegans spans cosmids U50184 and U00066 (GenBank TM /EBI accession numbers). The Arabidopsis thaliana homolog is on cosmid AL022537 (Gen-Bank TM /EBI accession number).

Isolation of the Human and Mouse HEL308
Genes-With the aim of isolating proteins involved in DNA cross-link repair, we searched data bases with the DNA helicase portion of Drosophila mus308. We found three human expressed sequence tags, AA625285, H08004, and R24580 (NCBI accession numbers), homologous to motif VI of the D. melanogaster mus308 helicase domain. From this ϳ700-bp sequence, primers were designed to clone the whole gene by 3Ј and 5Ј rapid amplification of cDNA ends (CLONTECH SMART RACE cDNA Amplification kit). Total RNA was prepared from a testis cancer cell line (833K) and a bladder cancer cell line (MGH-U1). Two different amplification reactions were performed for each cell line, and in both cases, two fragments of 1767 and 2167 bp were amplified through 3Ј RACE and 5Ј RACE, respectively. Six different PCR products (three from the 3Ј RACE and three from the 5Ј RACE) were sequenced to exclude mistakes obtained by PCR amplification. Finally, the complete open reading frame (3303 bp) was constructed by subcloning the 5Ј-and 3Ј-fragments into NcoI and KpnI sites of plasmid pFastBac HTb. To reflect the helicase activity described below and the high homology to mus308, we designated the previously unannotated gene as HEL308.
The human HEL308 gene maps on chromosome 4q21 and encodes for a protein of 1101 amino acids with a predicted molecular size of 124.5 kDa. We also found a mouse expressed sequence tag (NCBI accession number AA517170) that allowed us to clone mouse Hel308 by 3Ј and 5Ј rapid amplification of cDNA ends. Mouse Hel308 maps on chromosome 5-E in a region that is syntenic to human 4q21.
Human and mouse proteins share 88.6% identity and 93.5% similarity in the region shown in Fig. 1, and they belong to the "superfamily II" of DNA and RNA helicases. We aligned human and mouse HEL308 to D. melanogaster Mus308, C. elegans MUS-1 (3), human POLQ, and the A. thaliana homolog that we found in the data base (Fig. 1). These proteins share ϳ50% similarity, and they fall into the group of helicases designated by Harris et al. (11) as the MUS308 subfamily. Over the helicase domain region, Drosophila Mus308 has 40% identity (55% similarity) with HEL308 and 50% identity (66% similarity) with human POLQ. Human POLQ and human HEL308 share 40% identity (55% similarity) over the same region. No other human sequences were found approaching these high levels of homology to the Drosophila Mus308 helicase domain.
Purification of the Human HEL308 Protein-The HEL308 open reading frame was cloned into plasmid pFastBac HTb, linked to a hexahistidine tag at the N-terminal end. We also generated a cDNA expression construct encoding a tagged version of HEL308 with a single amino acid substitution at position 365. This conserved lysine residue is in the putative Walker A nucleotide binding motif and was changed to a methionine residue using site-directed mutagenesis. The resulting protein is referred to as HEL308 K365M . For a number of ATPases, including E. coli UvrD and S. cerevisiae Rad3, mutation of the equivalent lysine residue severely impairs nucleotide triphosphate hydrolysis, although the overall structure of the protein seems to remain intact (21)(22)(23).
For protein production, both cDNAs were placed under the transcriptional control of the polyhedrin promoter in recombinant baculoviruses. These viruses were used to transfect Sf9 cells. The proteins HEL308 and HEL308 K365M were detected in cell extracts by immunoblotting (data not shown). The extract was fractionated over a Ni 2ϩ -nitrilotriacetate superflow resin (Qiagen), a Mono Q column, and a Mono S column, and samples of the HEL308-containing fractions were analyzed by electrophoresis through an SDS-polyacrylamide gel (Fig. 2). A sample of the final purification step of the HEL308 K365M protein, which was produced and purified in exactly the same manner as the wild type protein, is shown in Fig. 2, lane 5.
ATPase and Helicase Activity of Human HEL308 -To characterize human HEL308, helicase and ATPase assays were performed on fractions along the Mono Q column gradient. Human HEL308 co-fractionated with a helicase activity and an ATPase activity (Fig. 3), though protein eluting after the peak fraction may be in a less active form. The helicase assay tested the ability to displace a 24-nt oligonucleotide from M13mp18 viral DNA (Fig. 3B). In the ATPase assay, released radiolabeled phosphate was separated from non-hydrolyzed ATP by thin layer chromatography, and the extent of hydrolysis was quantified (Fig. 3C). This ATPase activity was dependent upon the addition of single-stranded DNA to the reaction mixture, as found for other related enzymes (24).
Human HEL308 was further purified through a Mono S column. HEL308 ATPase activity was strongly stimulated by single-stranded DNA but not by double-stranded plasmid DNA (Fig. 4A). Unwinding of the 40-bp partial duplex substrate catalyzed by HEL308 required a nucleotide cofactor (Fig. 4B), as expected. Substituting 4 mM Mn 2ϩ instead of Mg 2ϩ gave barely detectable helicase activity (Fig. 4B, lane 1). A nonhydrolyzable ATP analog, AMP-PNP, could not support the helicase activity (Fig. 4B, lane 2). To determine whether nucle- Roman numerals indicate regions conserved in DNA and RNA helicases, and the bars mark their approximate extent. The arrow indicates the lysine that was mutated to methionine in mutant HEL308 K365M . Positions with six or more identical residues are shaded. The asterisks denote residues that are conserved in the MUS308 subfamily of helicases but not in other known or putative helicases. The sequence alignment was carried out using the Clustal X program. otide hydrolysis was needed for helicase activity, increasing amounts of AMP-PNP or ATP␥S were added to reaction mixtures that contained 2 mM ATP (Fig. 4C). AMP-PNP at 10 mM and ATP␥S at 2 mM were competitive inhibitors of helicase action, indicating that they bind to HEL308 in place of ATP and that nucleotide hydrolysis is necessary for activity. The result also suggests that ATP␥S has a higher affinity for HEL308 nucleotide binding sites than AMP-PNP. Drosophila RecQ5 helicase activity is similarly inhibited by an equimolar ratio of ATP␥S to ATP but not by an equimolar ratio of AMP-PNP to ATP (25). Interestingly, 2-4 mM AMP-PNP stimulated unwinding activity by about 2-fold in the presence of 2 mM ATP. This may suggest cooperative binding to an enzyme that has both catalytic and noncatalytic nucleotide binding sites, as found for the hexameric T7 gene 4 and E. coli Rho helicases (26,27). T7 gene 4 helicase prefers dTTP as a nucleotide cofactor, and the non-hydrolyzable nucleotide analog dTMP-PCP can bind to a noncatalytic site with little effect on dTTPase turnover (27).
To determine the nucleotide preference of HEL308, eight nucleotides were tested for their ability to support unwinding of the 40-bp partial duplex substrate. At the 2 mM nucleotide concentration used, only ATP and dATP gave detectable activity with 76% displacement in the presence of dATP and 33% displacement with ATP (Fig. 4B). ATP was used in further experiments since the ATP concentration in the cell is higher than the dATP concentration.
Unwinding of Longer DNA Duplexes by HEL308 -To determine whether HEL308 can unwind longer DNA duplexes, similar DNA substrates were constructed containing 60-or 70-nt fragments annealed to M13mp18. DNA helicase assays were performed with each DNA substrate in the presence of increasing amounts of HEL308 (Fig. 5). The mutant HEL308 K365M could displace none of the substrates (Fig. 5, lanes 3, A-F) and had no ATPase activity (not shown), confirming that these catalytic activities are due to wild type HEL308 enzyme. The HEL308 helicase displaced both the 60-and the 70-nt fragments, although with much less efficiency than the 20-and 40-nt fragments. After 30 min, 1.4 nM HEL308 displaced 89, 32, 2.6, and 2.5% of the 20-, 40-, 60-, and 70-nt oligonucleotides, respectively.
Substrates with 3Ј-or 5Ј-unpaired flaps were also tested. A non-complementary stretch of (A) 10 was added either 5Ј or 3Ј of the 40-nt fragment (Fig. 5, C and D). In neither case was unwinding activity changed, as compared with the displacement of the 40-nt fragment (Fig. 5G). We followed the unwinding reaction of the 20-bp partial duplex substrate at different time points using 0.14 nM HEL308 (Fig. 5H). After 10 min, 20% of the substrate was unwound with 50% unwinding reached after 18 min. The percentage of unwound substrate increased further with time.
Single-stranded DNA-binding Protein RPA Stimulates the HEL308 Helicase-One possible reason for the fact that the HEL308 helicase is less efficient in unwinding DNA duplexes of increasing length might be that the displaced single strand tends to re-anneal. If this is the case, the activity of HEL308 might be stimulated by single-stranded DNA-binding proteins. To test this possibility, the helicase activity was measured in the presence of purified human RPA (Fig. 6). In reactions containing the annealed 70-nt fragment and 2.8 nM HEL308, RPA stimulated HEL308 helicase activity with maximal stimulation at ϳ2 nM RPA. At this concentration, RPA increased displacement of the 70-nt fragment by 2.6-fold. 7.5 nM RPA did not stimulate HEL308 helicase activity (Fig. 6, lane 8), and concentrations of RPA equal or higher than 15 nM inhibited its activity (data not shown). These results suggest that RPA stimulates HEL308 by binding to the unwound regions produced by HEL308 helicase activity and inhibiting re-annealing. Given a binding site of about 30 nt for human RPA (28), 15 nM RPA covers approximately 25% of the M13 single-stranded DNA annealed to the 70-nt fragment. This concentration of RPA may more likely inhibit HEL308 translocation on DNA rather than its binding to DNA. The order of addition of RPA and HEL308 did not affect these results (data not shown).
The HEL308 Helicase Acts in the 3Ј to 5Ј Direction-To determine the polarity of the HEL308 helicase ,the 60-nt oligonucleotide was annealed to M13mp18 DNA, cut with SalI, and labeled at the 3Ј-ends with 32 P. This produced linear M13mp18 DNA with a 20-nt fragment annealed to its 5Ј-end and a 44-nt fragment annealed to its 3Ј-end. HEL308 displaced only the 20-nt fragment, indicating translocation in the 3Ј to 5Ј direction (Fig. 7A). Because HEL308 helicase activity decreases as a function of the size of the annealed oligonucleotide that it has to displace, a second DNA substrate was prepared. In this case, 42-and 32-nt fragments were annealed at the 5Ј-and 3Ј-end of linear M13mp18 DNA, respectively. HEL308 helicase activity displaced only the 42-nt fragment (Fig. 7B). These results confirm that the polarity of the HEL308 helicase is 3Ј to 5Ј relative to the single-stranded region.
Gel Filtration Analysis of Human HEL308 -The oligomeric structure of a DNA helicase is an important parameter that may have implications for its mechanism of action (29). The native molecular weight of human HEL308 was estimated from gel filtration analysis. Fraction 28 of the Mono Q column was loaded onto a Superose column equilibrated in buffer that included 0.5 M KCl. Proteins were eluted, and fractions were analyzed. Immunoblotting, ATPase, and helicase activities all co-eluted at a position corresponding to a molecular size of ϳ600 kDa as determined from standards (Fig. 8). This would be consistent with a possible hexameric association given the predicted molecular size of 124 kDa for a HEL308 monomer. Under the same gel filtration conditions, the mutant HEL308 K365M protein eluted at the same position as wild type HEL308 as determined by immunoblotting (not shown). This suggests that the point mutation K365M in the Walker A motif does not alter the quaternary structure of the protein. DISCUSSION We isolated human and mouse homologs of the helicase domain of D. melanogaster mus308 and designated them HEL308. Drosophila mus308 encodes a DNA polymerase in its C-terminal portion and a DNA helicase in its N-terminal portion (11). We found homologs of Drosophila mus308 in C. elegans and A. thaliana in addition to those in human and mouse, but we could not find any homolog in S. cerevisiae or in other yeast and bacterial species. The presumably full-length cDNA for human POLQ also has a predicted helicase domain at the N terminus, but its activity remains to be demonstrated.
The MUS308 family of helicases is part of the superfamily II of DNA and RNA helicases. Harris et al. (11) identified other sequences in GenBank TM that have sequence similarity to the helicase domain of Drosophila mus308. Besides the seven characteristic motifs of the superfamily II, these putative helicases include motifs Ib and IVa containing characteristic residues in the MUS308 subfamily. Motif I (the Walker A box) is unusual in having a serine in place of the glycine found in almost all Walker A motifs of known or putative helicases (Fig. 1, asterisk). Other residues, shown with an asterisk in Fig. 1, are peculiar to the MUS308 family of helicases: in motif V, the threonine is usually a hydrophobic residue in other members of superfamily II; in motif VI, the methionine is on the contrary usually a charged residue. Moreover, the MUS308 subfamily seems to combine motifs from various families. Motif II is characteristic of the so-called DEXH family of helicases, which includes many "repair/recombination" helicases, such as Rad3, Rad54, WRN, and BLM, whereas motif IV has characteristics of the DEAD family of RNA helicases.
HEL308 is a single-stranded DNA-dependent ATPase and DNA helicase. It translocates on DNA with 3Ј to 5Ј polarity and efficiently displaces 20-to 40-mer duplex oligonucleotides. Although activity on longer substrates is lower, it can be stimulated by the single-stranded DNA-binding protein RPA. Other helicases have been shown to be stimulated by single-stranded DNA-binding proteins. For example, S. cerevisiae MER3 helicase efficiently unwinds a 631-nt fragment only in the presence of RPA (30). Substrates with 3Ј-or 5Ј-unpaired flaps do not stimulate HEL308 activity, whereas many helicases, such as E. coli DnaB, phage T4 gene 41 protein, and phage T7 gene 4 protein, require a forked DNA substrate to initiate DNA unwinding in vitro (29). WRN helicase displaces a 40-nt oligonucleotide much more efficiently when two unpaired 10-mers are added to the 40-mer at its 3Ј-and 5Ј-ends (31). Both BLM and WRN prefer a fork DNA substrate to unwind DNA (32). Gel filtration analysis suggests that HEL308 behaves as a multimer, possibly a hexamer. A few DNA helicases studied to date operate as monomers, such as PcrA (33) (29,35). Studies by electron microscopy will reveal whether HEL308 forms multimeric structures.
Drosophila mus308 is believed to be involved in the repair of interstrand DNA cross-linking damage since mus308 mutants are hypersensitive to DNA cross-linking agents such as photoactivated psoralen, diepoxybutane, and nitrogen mustard but are not sensitive to the monofunctional alkylating agent methyl methanesulfonate (11). Not much is known about how interstrand cross-links are repaired in eukaryotes. Biochemical and genetic analyses in prokaryotes indicate that DNA interstrand cross-links are repaired by an excision-recombination mechanism (36,37). In E. coli, removal of a DNA interstrand cross-link is initiated by the UvrABC endonuclease complex, which incises one strand on each side of the cross-link. The 5Ј nuclease activity of DNA polymerase I, in concert with the UvrD helicase, generates a gap at the site of incision. The resulting single-stranded region provides a substrate for binding of RecA protein and for the initiation of homologous pairing and strand exchange. The polymerase activity of DNA polymerase I can then carry out repair synthesis with an undamaged homolog as a template. Some aspects of this mechanism of interstrand cross-link repair may be conserved in eukaryotes. Genetic data suggest that both excision and recombination are involved in the repair of interstrand cross-links in eukaryotes as well. Mus308 and POLQ belong to the A family of DNA polymerases, as does E. coli DNA polymerase I. Both UvrD and HEL308 are 3Ј to 5Ј multimeric DNA helicases. This direction of translocation along DNA would fit a model where coupling between a helicase and a polymerase coordinates duplex unwinding and polymerization. Gene 4 of bacteriophage T7 encodes a protein (gp4) that has both a helicase and an RNA primase domain. gp4 forms hexameric rings that can translocate along single-stranded DNA, coupling the unwinding of duplex DNA with the synthesis of short RNA primers that are elongated by T7 DNA polymerase (38). It will be interesting to analyze whether HEL308 interacts with a polymerase and forms a complex with an activity similar to T7 gp4. The existence of both HEL308 and the very similar putative helicase of POLQ raises further questions concerning the functions of these enzymes. It is possible that they function in DNA repair processes in different tissues.
Fanconi anemia cell lines are sensitive to interstrand crosslinking agents. Thus far, eight Fanconi anemia complementation groups have been defined, and six genes have been mapped and cloned (1). We are currently investigating the possibility , and percentage of released phosphate (ATPase assay) for each fraction. Protein concentration on the immunoblot was quantified with the ChemiDoc chemiluminescence detection system (Bio-Rad Laboratories). Displaced radioactivity was quantified with the Fuji phosphorimaging device.