GENOMICS 54, 424–436 (1998) ARTICLE NO. GE985587

cDNA Cloning and Mapping of Human Homologs for Schizosaccharomyces pombe , rad1, and and Cloning of Homologs from Mouse, Caenorhabditis elegans, and Drosophila melanogaster

Frank B. Dean,*,1 Lubing Lian,† and Mike O’Donnell*,‡

*The Rockefeller University and ‡The Howard Hughes Medical Institute, 1230 York Avenue, New York, New York 10021; and †Myriad Genetics, Inc., 390 Wakara Way, Salt Lake City, Utah 84108

Received July 13, 1998; accepted September 17, 1998

INTRODUCTION Mutations in DNA repair/cell cycle checkpoint can lead to the development of cancer. The cloning of The progression of a eukaryotic cell through the human homologs of yeast DNA repair/cell cycle check- stages of the cell cycle can be arrested if the events of point genes should yield candidates for human tumor the previous stage of the cell cycle, such as DNA rep- suppressor genes as well as identifying potential tar- lication, have not been completed or if the DNA has gets for cancer therapy. The Schizosaccharomyces sustained some type of damage. The controls on cell pombe genes rad17, rad1, and hus1 have been identi- cycle progression are termed checkpoints (Hartwell fied as playing roles in DNA repair and cell cycle and Weinert, 1989), and they are able to detect checkpoint control pathways. We have cloned the whether the processes of the individual stages of the cDNA for the human homolog of S. pombe rad17, cell cycle have been completed and whether the DNA is RAD17, which localizes to chromosomal location 5q13 intact or in need of repair. Cells that are mutated in by fluorescence in situ hybridization and radiation one of the cell cycle checkpoint genes, however, are able hybrid mapping; the cDNA for the human homolog of to proceed from one stage of the cell cycle to the next S. pombe rad1, RAD1, which maps to 5p14–p13.2; and even if the cellular processes of that stage are incom- the cDNA for the human homolog of S. pombe hus1, plete or in the presence of DNA damage. The G2 phase HUS1, which maps to 7p13–p12. The human gene loci of the cell cycle lies between S phase, in which DNA have previously been identified as regions containing replication takes place, and M phase, when mitosis tumor suppressor genes. In addition, we report the occurs. Thus the G2 checkpoint is critical for ensuring cloning of the cDNAs for genes related to S. pombe that mitosis does not occur until the necessary steps of rad17, rad9, rad1, and hus1 from mouse, Caenorhab- DNA replication, DNA repair, and dupli- ditis elegans, and Drosophila melanogaster. These in- cation are complete. clude Rad17 and Rad9 from D. melanogaster, hpr-17 Many checkpoint-deficient mutants have been iden- and hpr-1 from C. elegans, and RAD1 and HUS1 from tified in the budding yeast Saccharomyces cerevisiae mouse. The identification of homologs of the S. pombe and in the fission yeast Schizosaccharomyces pombe. rad checkpoint genes from mammals, arthropods, and Genes that link mitosis to the completion of DNA rep- nematodes indicates that this cell cycle checkpoint lication have been isolated (Enoch and Nurse, 1990; pathway is conserved throughout eukaryotes. © 1998 Enoch et al., 1992; McFarlane et al., 1997). In addition, Academic Press many genes that function in DNA repair have been identified as G2 checkpoint control genes (Nasim and Smith, 1975; Al-Khodairy and Carr, 1992; Al-Khodairy Sequence data from this article have been deposited with the et al., 1994), including S. cerevisiae RAD9 (Weinert and EMBL/GenBank Data Libraries under Accession Nos. AF076838, Hartwell, 1990), S. cerevisiae MEC3 (Weinert et al., AF076839, AF076840, AF076841, AF076842, AF076843, AF076844, 1994), S. pombe rad1 (Rowley et al., 1992), S. pombe AF076845, AF076846, AF090170, G41776, G41777, and G41778. rad3 (Jimenez et al., 1992; Bentley et al., 1996), S. 1 To whom correspondence should be addressed at The Rockefeller University, 1230 York Avenue, Box 228, New York, NY 10021. Tele- pombe rad9 (Murray et al., 1991), S. pombe rad17 phone: (212) 327-7255. Fax: (212) 327-7253. E-mail: deanfb@mod. (Griffiths et al., 1995), S. pombe hus1 (Kostrub et al., rockefeller.edu. 1997), and the fungus Ustilago maydis REC1 (Onel et

424 0888-7543/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved. HUMAN RAD CHECKPOINT HOMOLOGS 425 al., 1996). The S. pombe rhp9 gene is a homolog of S. pombe rad3 (Savitsky et al., 1995; reviewed by Enoch cerevisiae RAD9 (Willson et al., 1997). Recent studies and Norbury, 1995; Lehmann and Carr, 1995; Jackson, have begun to indicate what the in vivo role of the S. 1996). These have kinase activity and cerevisiae checkpoint genes may be (Lydall and Wein- are involved in generating a signal to halt progression ert, 1995). The Rad24, Rad17, and Mec3 proteins ap- through the cell cycle in response to DNA damage. pear to activate an exonuclease activity in vivo while Identification of the human homologs of the yeast G2 the Rad9 protein appears to modulate exonuclease ac- cell cycle checkpoint genes should yield attractive can- tivity. A number of reviews summarize this work (Shel- didates for novel human tumor suppressor genes. The drick and Carr, 1993; Lydall and Weinert, 1996; Stew- human genes are likely to play a cell cycle regulatory art and Enoch, 1996; Carr, 1997). role in human cells, and previously identified human The gene for S. pombe rad17 has been described cell cycle checkpoint genes have been identified as tu- (Griffiths, 1995), as has the gene for its homolog in S. mor suppressors. A human homolog of S. pombe rad9 cerevisiae, RAD24 (Lydall and Weinert, 1997). Cloning was described recently (Lieberman et al., 1996). of S. pombe rad17 revealed that it has an ATP binding We describe here the cDNA cloning of the human site and has extensive homology over its entire length homologs of the S. pombe rad17, rad1, and hus1 genes. to the DNA polymerase accessory proteins known as We have mapped the chromosomal location of the hu- clamp loaders (Griffiths et al., 1995). Clamp loaders man genes and find that the map positions of RAD17 (for instance human RFC, Escherichia coli ␥ complex) and RAD1 correlate with loci for human tumor sup- couple hydrolysis of ATP to positioning a protein ring pressor genes. In addition, we have cloned cDNAs for around duplex DNA. The protein ring tethers the rep- some of the homologs of these genes from mouse, Dro- licase to DNA for high processivity in chromosomal sophila melanogaster, and Caenorhabditis elegans. Re- replication (Kelman and O’Donnell, 1995). The homol- ports of the cloning of human RAD17, RAD1, and ogy to clamp loading subunits suggests that S. pombe HUS1 appeared while this work was in progress rad17 may carry out a similar clamp loading or unload- (Parker et al., 1998a, b; Kostrub et al., 1998). ing function in the DNA repair pathway. Interestingly, while the S. pombe rad17 gene carries out two roles, DNA repair and cell cycle checkpoint regulation, the MATERIALS AND METHODS two functions are separable. Specific point mutations of Isolation of human RAD17. Two sequences were identified in the rad17 that reduced DNA repair activity but did not EST database for human cDNA clones (Accession No. T10666, clone affect checkpoint control were generated. The cloning hbc863, and Accession No. AA133547, clone 586844). Clone hbc863 of the gene for S. pombe rad1 has been described in a was kindly sent by Graeme Bell, The University of Chicago. Clone pair of reports (Sunnerhagen et al., 1990; Long et al., 586844 (IMAGE, Integrated Molecular Analysis of Genomes and Their Expression) was obtained from Genome Systems, Inc. Rapid 1994). The DNA repair and checkpoint functions of amplification of cDNA ends was carried out using Marathon-Ready rad1 are also separable; mutations that have differen- cDNA (Clontech, Palo Alto, CA) as a template. Primer R1702 tial effects on the two activities have been isolated (5Ј-GCGGGATCCCTATGTCCCATCACTCTCGTAGTCTTC-3Ј) an- (Kanter-Smoler et al., 1995). The cloning of its S. cer- nealed to the 3Ј end of the open reading frame and primed synthesis evisiae homolog, RAD17, has also been described of the antisense strand. Primer AP1 (Clontech) annealed to the Marathon cDNA adaptor product strand at the 5Ј end of the cDNA (Siede et al., 1996). Extensive work has been performed and primed synthesis of the sense strand. The PCR amplification on its homolog in U. maydis, REC1 (Holliday et al., was performed for 30 cycles of 94°C for 30 s and 68°C for 7 min in a 1976; Holden et al., 1989; Tsukuda et al., 1989). An reaction volume of 50 ␮l containing 40 mM Tricine–KOH, pH 9.2 at exonuclease activity is associated with the protein 25°C, 15 mM KOAc, 3.5 mM Mg(OAc)2,75␮g/ml bovine serum (Thelen et al., 1994), and, again, the gene plays roles in albumin, 200 ␮M each dNTP, 10 pmol each primer AP1 and R1702, 0.5 ng Marathon-Ready cDNA template, 1 ␮l Advantage KlenTaq both DNA repair and cell cycle regulation (Onel et al., Polymerase Mix (Clontech). Amplified DNA products (15 ␮l) were 1995, 1996). The cloning of S. pombe hus1 was de- analyzed by electrophoresis through a 1.0% agarose gel, and a band scribed recently (Kostrub et al., 1997). Yeast strains of DNA 2.6 kb in size was observed. A second PCR amplification was disrupted in hus1 are viable, but are checkpoint-defec- performed under conditions identical to those of the first reaction, tive. with the exception that the template used was 0.5 ng of PCR product of the first reaction, and 10 pmol of primer AP2 (Clontech) was used The occurrence of mutations in the checkpoint con- instead of primer AP1. Amplified DNA products were separated by trol genes of higher eukaryotes can lead to cancer electrophoresis through a 1.0% low-melting agarose gel, and a gel (Hartwell, 1992; Hartwell and Kastan, 1994; Kastan, slice containing the 2.6-kb reaction product was excised from the gel. 1991; Kuerbitz et al., 1992). Genes that, when mutated, The 2.6-kb PCR product was cloned using the TOPO TA Cloning vector (Invitrogen, Carlsbad, CA). DNA minipreps (48) were carried allow increased rates of tumor formation are termed out, yielding 27 clones containing the PCR product. Five clones were tumor suppressors. Many tumor suppressors have cell sequenced, and 3 were found to contain the complete, identical, open cycle checkpoint function, and loss-of-function muta- reading frame. The gene was designated RAD17, and the cDNA tions in these genes causes runaway cell proliferation, sequence has been deposited with the GenBank database under leading to tumor formation (Collins et al., 1997). For Accession No. AF076838. Gene-specific primers R1715 (5Ј-ACCTG- TATACCTTTTGAAACGTCACAG-3Ј) and R1702 were used to am- example, ATM has been identified as the gene that is plify the human RAD17 cDNA from Marathon-Ready cDNA in a PCR defective in patients with ataxia telangiectasia. ATM is as described above. The 2.6-kb reaction product was purified using a human homolog of S. cerevisiae MEC1/ESR1 and S. the QIAquick PCR purification kit (Qiagen) and sequenced directly. 426 DEAN, LIAN, AND O’DONNELL

Isolation of D. melanogaster Rad17. An EST specifying a cDNA the BioNick labeling kit (GIBCO BRL) for use as a probe. The probes from D. melanogaster (Accession No. AA539148, clone LD17237) was hybridized to metaphase of human lymphocytes with identified. Clone LD17237, from the Berkeley Drosophila Genome an efficiency of between 67 and 75%. Project/Howard Hughes Medical Institute (BDGP/HHMI) Drosoph- For radiation hybrid mapping (Cox et al., 1990) of HUS1, RAD1, ila EST Project, was obtained from Genome Systems, Inc. The DNA and RAD17, sequence tagged sites (STSs) were developed from the 3Ј was sequenced, and it appeared to encode a full-length cDNA. The ends of each of the genes. For HUS1, the primers CCTGCGCTGTC- gene was designated Rad17, and the cDNA sequence has been de- CTAGCACCCT and ATCCAACTGTGTGTTGTTGAATC were used posited with the GenBank database under Accession No. AF076839. to amplify a 187-bp product (GenBank Accession No. G41778). For Rad1, the primers GTTCCTGAATCTGAGTCTTGAG and CTC- Isolation of C. elegans hpr-17. A sequence specifying a cDNA CTATCTTCCCCATTGTGCT were used to generate a 148-bp product from C. elegans (Accession No. D75465, clone yk104h11) was identi- (GenBank Accession No. G41777). In the case of RAD17, the primers fied in the EST database. Clone yk104h11 was kindly sent by Yuji ACATAGAAGCCAGCCTGCTAAT and AGAAGAGTAGGTTTCTT- Kohara, National Institute of Genetics, Japan. The DNA sequence GTGATG generated a 161-bp product (GenBank Accession No. was determined, the gene was designated hpr-17 for “homolog of S. G41776). Taq Platinum polymerase (Gibco) and the following PCR pombe rad,” and the cDNA sequence has been deposited with the conditions were used to amplify the three STSs: 37 cycles of 96°C for GenBank database under Accession No. AF076840. 15 s, 60°C for 20 s, and 72°C for 40 s. Under these conditions, the Isolation of human RAD1. Two ESTs for human cDNA clones STSs were found to amplify specifically the expected-sized product were identified (Accession No. AA227739, clone 667461 and Acces- from human genomic DNA but not hamster genomic DNA. The sion No. AA029300, clone 470124). Both clones (IMAGE) were ob- Genebridge4 somatic hybrid cell panel was used to map the genomic tained from Genome Systems, Inc. The gene was designated RAD1, locations of the three genes (Gyapay et al., 1996). and the two cDNA sequences have been deposited with the GenBank database under Accession Nos. AF076841 and AF090170. Gene- specific primers R105 (5Ј-AGGGAATTCCATATGCCCCTTCTGAC- RESULTS CCAACAGATCCAA-3Ј) and R106 (5Ј-GCGGGATCCTCAAGACTCA- Isolation and Sequence Comparison of rad17 GATTCAGGAACTTCTTC-3Ј) were used to amplify the open reading frames for the two human RAD1 cDNAs from Marathon-Ready Homologs cDNA as described above. A BLAST search of the GenBank EST database Isolation of mouse RAD1. An EST for a mouse cDNA clone was shows a partial human cDNA sequence with a similar- identified (Accession No. AA387463, clone 789687). Clone 789687 (IMAGE) was obtained from Genome Systems, Inc. The gene was ity between its translated amino acid sequence and the designated RAD1, and the cDNA sequence has been deposited with S. pombe rad17 amino acid sequence. The cDNA clone, the GenBank database under Accession No. AF076842. EST59509, was acquired from The Institute for Ge- Isolation of C. elegans hpr-1. An EST specifying a cDNA from C. nome Research, and its DNA sequence was deter- elegans (Accession No. D76378, clone yk117e8) was identified. Clone mined. However, it was found not to encode a full- yk117e8 was kindly sent by Yuji Kohara, National Institute of Ge- length cDNA molecule. The database of ESTs at the netics, Japan. The gene was designated hpr-1, and the cDNA se- National Institutes of Health was then searched with quence has been deposited with the GenBank database under Acces- the translated amino acid sequence of the partial sion No. AF076843. cDNA using the National Center for Biotechnology In- Isolation of human HUS1. Two ESTs for human cDNA clones formation BLAST server and the TBLASTN program were identified (accession AA280710, clone 711713, and accession R29753, clone F1-1279D). Clone 711713 (IMAGE) was obtained from (Altschul et al., 1990). Two ESTs that identified poten- Genome Systems, Inc. Clone F1-1279D was kindly sent by Sun Shim tially full-length human cDNA clones were found, and Choi and Hee-Sup Shin, Pohang Institute of Science & Technology, the DNA sequences of the two clones were determined. Korea. The gene was designated HUS1, and the cDNA sequence has The sequences of the three clones overlapped one an- been deposited with the GenBank database under Accession No. other and were assembled into one cDNA sequence AF076844. Gene-specific primers H121 (5Ј-AGGGAATTCCATAT- GAAGTTTCGGGCCAAGATCGTGGAC-3Ј) and H122 (5Ј-GCGG- containing an open reading frame of about 2000 nucle- GATCCCTAGGACAGCGCAGGGATGAAATACTG-3Ј) were used to otides, which appeared long enough to encode a protein amplify the human HUS1 open reading frame from Marathon-Ready of the expected size. However, the cDNA sequence was cDNA as described above. not complete because no methionine initiator codon Isolation of mouse HUS1. Two ESTs for mouse cDNA clones were was present in the 5Ј portion of the open reading frame. identified (Accession No. AA153060, clone 604141, and Accession No. A full-length cDNA was isolated via the technique of AA218365, clone 658994). Both clones (IMAGE) were obtained from rapid amplification of cDNA ends, and the gene was Genome Systems, Inc. The gene was designated HUS1, and the cDNA sequence has been deposited with the GenBank database designated RAD17. Interestingly, the 5Ј untranslated under Accession No. AF076845. sequence of the cDNA appeared to be extremely unsta- ble in E. coli. Of 27 independently isolated recombinant Isolation of D. melanogaster Rad9. A sequence specifying a cDNA from D. melanogaster (Accession No. AA391137, clone plasmids carrying the cDNA insert, 24 carried unique LD10092) was identified. Clone LD10092, from the BDGP/HHMI deletions near the 5Ј end. This was true even for re- Drosophila EST Project, was obtained from Genome Systems, Inc. combinant plasmids isolated from an E. coli strain The gene was designated Rad9, and the cDNA sequence has been containing mutations in several recombination genes deposited with the GenBank database under Accession No. (SURE2 cells, Stratagene; recB recJ sbcC umuC::Tn5 AF076846. uvrC). Plasmids containing the RAD17 open reading Mapping of the RAD genes on human chromosomes. Fluorescence frame as an insert but lacking the 5Ј untranslated in situ hybridization (FISH) was carried out by SeeDNA Biotech, Inc. (Ontario, Canada) (Heng et al., 1992; Heng and Tsui, 1993). Human sequence were stable. To determine the cDNA se- RAD17 cDNA (clone 586844, 2.5 kb), RAD1 cDNA (clone 667461, 1.2 quence more directly, the PCR product amplified from kb), or HUS1 cDNA (clone F1-1279D, 1.8 kb) was biotinylated using cDNA was sequenced. Its sequence confirmed the DNA HUMAN RAD CHECKPOINT HOMOLOGS 427 sequence determined from the cloned cDNA products. quences of both clones were determined. Clone 667461 Its open reading frame specified a gene product of 670 contained a full-length open reading frame, while clone amino acids, with a predicted molecular weight of 76 470124 contained a deletion of 109 nt that disrupted kDa. the amino acid coding region. We tested whether the The cDNAs for rad17 homologs from D. melano- clone carrying the deletion represented a cloning arti- gaster and C. elegans were identified using the data- fact or a bona fide species of human mRNA. Two cDNA base of ESTs and the human Rad17p amino acid se- preparations, one derived from human fetal brain and quence. The two cDNA clones were acquired and one from HeLa cells, were analyzed by PCR for the sequenced. The fly Rad17 gene product is 520 amino presence of RAD1 cDNAs. Surprisingly, two amplifica- acids long, with a molecular mass of 57 kDa. The open tion products were generated from each cDNA prepa- reading frame of the C. elegans homolog encoded 514 ration. They were of the expected 870 and 760 bp sizes, amino acids but appeared to be incomplete, lacking in an approximately 60:40 ratio (data not shown). The perhaps 5% of the hpr-17 open reading frame at its 5Ј two cDNA molecules appear to be derived from the end. same gene, because their nucleotide sequences are Significant amino acid sequence identity and simi- identical, including the sequences of both the 5Ј and larity were observed between the S. pombe Rad17p and the 3Ј untranslated regions, and radiation hybrid map- the human, D. melanogaster, C. elegans, and budding ping indicates the presence of just a single gene locus yeast homologs. An alignment of the five predicted (see below). Therefore we suggest that the two cDNA amino acid sequences is shown in Fig. 1. A region that molecules represent RAD1 mRNA splicing variants of conforms to the P-loop ATP/GTP binding consensus a single gene. The same methionine codon is expected sequence (Koonin, 1993) was found in the N-terminal to be utilized for the initiation of protein synthesis from region of human RAD17 (residues 126 to 133), as well either mRNA, but the sizes of the two predicted protein as in the other four sequences (Griffiths, 1995). How- products are quite different. The predicted protein ever, the yeast sequence, instead of the highly con- product of the longer RAD1 open reading frame is 282 served GKS/T motif, carries the sequence SKS. The amino acids long, with a molecular mass of 32 kDa. The human, fly, and worm amino acid sequence at the predicted protein product of the open reading frame Walker B motif is EDxx, rather than DExx. The corre- with the 109-nt deletion consists of the amino-terminal sponding sequence in the fission yeast gene is TELP. 68 amino acids of the larger protein, with a correspond- The gapped BLASTP E values (Altschul et al., 1997) ing molecular mass of only 7.5 kDa. Whether such a for pairwise alignments between the S. pombe protein protein is produced in vivo has yet to be investigated. and the human, D. melanogaster, and C. elegans pro- The database of ESTs was then searched with the teins are also an indication that the sequences are amino acid sequence of human RAD1 using TBLASTN. significantly related (Table 1). The similarity between A full-length mouse clone was identified and se- the C. elegans and the human sequences is also high; quenced, and it encoded a full-length cDNA. The mouse the BLAST E value for the amino acid sequence align- RAD1 protein sequence is 280 amino acids long, also of ment is 2 ϫ 10Ϫ29. Interestingly, however, the se- molecular mass 32 kDa. The database of ESTs was quence similarity between the C. elegans and the S. again searched with the amino acid sequence of human pombe proteins is not so high (E value ϭ 4 ϫ 10Ϫ9). RAD1 using TBLASTN, and a C. elegans clone was Whether these genes are orthologs of one another is identified and sequenced. It also appeared to encode a perhaps better addressed by whether they are the most full-length cDNA. The C.elegans hpr-1 open reading similar proteins in the respective organisms. The genes frame is 267 amino acids in length, molecular mass 30 were originally identified as the ones in the database kDa. most similar to the S. pombe genes. The reverse is also Significant amino acid sequence identity and simi- true; S. pombe rad17 is the S. pombe gene in the larity were observed between the S. pombe Rad1p; the database that is most similar to the other eukaryotic human, mouse, and C. elegans homologs; the S. cerevi- homologs (Table 1). A caveat to these considerations is siae RAD17; and the U. maydis REC1. The E values of that the representation of human, S. pombe, C. el- the gapped BLASTP program for pairwise alignments egans, and D. melanogaster amino acid sequences in between the S. pombe protein and the human, mouse, the databases is not yet complete. It remains possible and C. elegans proteins showed that the sequences that proteins with greater sequence similarity will be were significantly related (Table 1). Again, S. pombe identified in the future. Rad1p is the S. pombe protein most similar to the human, mouse, and worm homologs. An alignment of Isolation and Sequence Comparison of rad1 the six predicted amino acid sequences is shown in Fig. Homologs 2. The YxGxGxPxxxxxE motif (Siede et al., 1996) was found in human RAD1 (residues 112 to 124), as well as The database of ESTs at the National Institutes of in the other five sequences, with the exception of one Health was searched with the amino acid sequence of amino acid change in the C. elegans sequence. S. pombe rad1 using TBLASTN. Two potentially full- U. maydis REC1 has an associated exonuclease ac- length human clones were identified, and the se- tivity (Thelen et al., 1994). Although a BLAST search 428 DEAN, LIAN, AND O’DONNELL

FIG. 1. Multiple alignment of the rad17 homologs. Human, Homo sapiens; Drome, D. melanogaster; Caeel, C. elegans; Pombe, S. pombe (Griffiths et al., 1995); Yeast, S. cerevisiae (Lydall and Weinert, 1997). Protein sequences were aligned by the ClustalW method using the Megalign program (DNASTAR, Inc., WI). Identical amino acids are indicated by black boxes while similar amino acids are shaded gray. Similar amino acids are M, I, L, and V; F, W, and Y; A and G; D and E; N and H; S and T. using the REC1 amino acid sequence does not identify mains I, II, and III (Thelen et al., 1994). However, the other nucleases as having sequence similarity, an highly conserved DxE motif of exonuclease domain I alignment between the E. coli polA, dnaQ, and REC1 (Koonin, 1997) is only weakly represented among the amino acid sequences indicated which REC1 amino Rad1 homologs (human RAD1 amino acids 77–78, acids might correspond to conserved exonuclease do- REC1 amino acids 147–149; see Fig. 2). HUMAN RAD CHECKPOINT HOMOLOGS 429

TABLE 1 Isolation and Sequence Comparison of hus1 Amino Acid Sequence Comparison between S. pombe Homologs Rad Checkpoint Genes and Their Homologs Two ESTs for the human homologs of S. pombe hus1 % Identity/% Gapped BLAST Similarity were identified in the database using the amino acid Gene similaritya E valueb to S. pombec sequence of S. pombe hus1 and the gapped TBLASTN program. The two clones were acquired, and their DNA Ϫ34 Human RAD17 23/41 3 ϫ 10 ϩ sequences were determined. Neither clone encoded a Fly Rad17 23/44 6 ϫ 10Ϫ21 ϩ Worm hpr-17 21/40 4 ϫ 10Ϫ9 ϩ full-length open reading frame. However, the two Human RAD1 29/53 2 ϫ 10Ϫ27 ϩ clones overlapped each other, having 259 nt in com- Mouse RAD1 30/54 1 ϫ 10Ϫ27 ϩ mon, thus identifying the full-length cDNA sequence. Ϫ11 Worm hpr-1 24/45 2 ϫ 10 ϩ Nucleotides 1850–2100 of the 3Ј untranslated region Human HUS1 30/54 2 ϫ 10Ϫ28 ϩ 29 appear to correspond to a member of the Alu repeat Mouse HUS1 30/55 1 ϫ 10Ϫ ϩ Human RAD9 22/43 6 ϫ 10Ϫ23 ϩ family. PCR was used to amplify the full-length open Fly Rad9 22/43 6 ϫ 10Ϫ15 ϩ reading frame from cDNA, and the 865 nucleotide-long reaction product was cloned. The sequence of the a Similar amino acids are defined in the legend to Fig. 1. cloned PCR product confirmed the data derived from b E value for the alignment between the S. pombe protein and its homolog. the two partial cDNA clones. The human HUS1 open c A“ϩ” indicates that the S. pombe protein is the most similar of all reading frame is 280 amino acids, of 32 kDa molecular S. pombe proteins in the database to the homolog. mass.

FIG. 2. Multiple alignment of the rad1 homologs. Human, H. sapiens; Mouse, Mus musculus; Caeel, C. elegans; Pombe, S. pombe (Long et al., 1994); Ustma, U. maydis (Onel et al., 1995); Yeast, S. cerevisiae (Siede et al., 1996). Amino acid sequences were aligned by the ClustalW method using Megalign. Amino acids 81 to 145 (indicated by **1**) and 191 to 135 (indicated by **2**) of the U. maydis REC1 sequence and the last 83 amino acids of the yeast amino acid sequence are omitted from the alignment. Identical amino acids are shown in black, and conservative substitutions are shown in gray. 430 DEAN, LIAN, AND O’DONNELL

FIG. 3. Multiple alignment of the hus1 homologs. Human, H. sapiens; Mouse, M. musculus; Pombe, S. pombe (Kostrub et al., 1997). Alignments were performed by the ClustalW method using Megalign. Identities with the S. pombe amino acid sequence are indicated with black boxes, while similarities are indicated by gray.

The database of ESTs was searched with the amino Mapping of RAD17, RAD1, and HUS1 on Human acid sequence of human HUS1 using TBLASTN. Two Chromosomes ESTs for a mouse homolog were identified and se- Chromosome mapping of human RAD17, RAD1, and quenced. Again, neither clone encoded a full-length HUS1 was performed by FISH and radiation hybrid open reading frame. However, the two clones over- mapping analyses. The biotinylated FISH probe of lapped each other, having 108 nt in common, and de- RAD17 hybridized to metaphase chromosomes of hu- fined a complete open reading frame. The mouse HUS1 man lymphocytes with an efficiency of 73% (among 100 predicted protein sequence is 281 amino acids long, of checked mitotic figures, 73 showed signals on one pair 32 kDa molecular mass. of the chromosomes). Based on 10 independent signals, Significant amino acid sequence identity and simi- RAD17 is located on human 5q13 (Fig. 5A). No existing larity were observed between the S. pombe Hus1p and STSs that correspond to RAD17, RAD1, or HUS1 were the human and mouse homologs. The gapped BLASTP present in the NCBI database. Therefore, to carry out E values for pairwise alignments between the S. pombe radiation hybrid mapping we generated STSs for these protein and the human and mouse proteins showed genes from their 3Ј untranslated regions and have that the sequences were closely related (Table 1). S. deposited them in GenBank (see Materials and Meth- pombe Hus1p is again the S. pombe protein most sim- ods). Using the Genebridge4 somatic hybrid cell panel ilar to the human and mouse proteins. An alignment of and an STS in the 3Ј region of RAD17, we found that the three amino acid sequences is shown in Fig. 3. the gene maps 246.1 cR from the p telomere of human , according to the Whitehead RH map location (MIT). Specifically, RAD17 maps 14 cR telo- Isolation and Sequence Comparison of rad9 meric of WI-3133 (alias D5S1872), which is present on Homologs the Whitehead WC5.6 physical mapping contig. This A human homolog of S. pombe rad9 has been iden- corresponds to an LDB chromosome map location of tified, as has a yeast homolog, DDC1 (Lieberman et al., 5q11–q13. 1996; Longhese et al., 1997). The database of ESTs was In the case of RAD1, the corresponding FISH probe searched with the amino acid sequence of human hybridized to metaphase chromosomes of human lym- phocytes with an efficiency of 75%. Based on 10 inde- RAD9 using TBLASTN, and an EST for a D. melano- pendent signals, RAD1 is located within human 5p14– gaster homolog was identified. The cDNA clone was p13.2 (Fig. 5B). One additional locus was detected on acquired and sequenced and appeared to encode a full- chromosome 10, region q25, at a frequency of 45%. length cDNA. The fly Rad9 predicted gene product is Since this frequency was lower than that on chromo- 456 amino acids long with a molecular mass of 51 kDa. some 5, we infer that RAD1 resides on chromosome 5 There is also significant amino acid sequence iden- while a related sequence, perhaps a pseudogene, maps tity and similarity between the S. pombe Rad9p and to 10q25. This hypothesis is supported by radiation the human, fly, and budding yeast homologs. The hybrid mapping data generated using an STS devel- gapped BLASTP E values for alignments between the oped from the 3Ј region of RAD1, which places the gene S. pombe protein and the human and fly proteins at 137.3 cR from the p telomere of human chromosome shows a degree of sequence similarity in line with that 5. Specifically, RAD1 maps 3.2 cR centromeric of seen with the other S. pombe rad gene homologs (Table AFM303TH5, which is found on the Whitehead WC5.3 1). A multiple alignment among S. pombe rad9 ho- physical mapping contig, and corresponds to an LDB mologs from human, fly, and yeast is shown in Fig. 4. map location of 5p13–p11. HUMAN RAD CHECKPOINT HOMOLOGS 431

FIG. 4. Multiple alignment of rad9 homologs. Human, H. sapiens (Lieberman et al., 1996); Drome, D. melanogaster; Pombe, S. pombe (Murray et al., 1991); Yeast, S. cerevisiae (Longhese et al., 1997). The alignment was performed by the ClustalW method using Megalign. Black boxes show identities, and gray boxes indicate similarities.

Finally, a FISH HUS1 probe hybridized to meta- are both present together on chromosome 5 and con- phase chromosomes of human lymphocytes with an firmed that RAD17 maps to chromosome 5p13 (Fig. 6). efficiency of 67%. Based on 10 independent signals, HUS1 is localized on human 7p13–p12 (Fig. 5C). Using DISCUSSION an STS developed from the 3Ј region of HUS1, the gene was radiation hybrid mapped to 182.5 cR from the p The cloning of rad/checkpoint gene homologs from telomere of , 5 cR centromeric of fungi, nematodes, arthropods, and vertebrates demon- CHLC.GATA24D12 (alias D7S1818), which is present strates that this G2 cell cycle checkpoint is highly on the Whitehead WC7.3 contig. The corresponding conserved among the various evolutionary lineages of LDB map location is 7p13–p11. eukaryotes. The rad/checkpoint genes of S. cerevisiae, Parker et al. (1998b) reported by FISH mapping that S. pombe, and humans involved in the G2 cell cycle human RAD17 localized to chromosome 4, region checkpoint are summarized in Table 2. The six rad q13.3–q21.2. To clear up this discrepancy SeeDNA Bio- checkpoint genes of S. pombe are rad17, rad1, hus1, tech generously repeated the RAD17 FISH mapping rad9, rad26 (Al-Khodairy et al., 1994), and rad3. At analysis by carrying out simultaneous hybridization of this point in time no homologs of S. pombe rad26 have human RAD17 and RAD1 probes to metaphase chro- been identified. The human gene ATM was originally mosomes of human lymphocytes. The FISH signals and identified as the homolog of S. cerevisiae MEC1 and S. DAPI banding pattern showed that RAD17 and RAD1 pombe rad3. It now seems likely that ATR is the hu- 432 DEAN, LIAN, AND O’DONNELL

FIG. 5. FISH mapping results for human RAD17, RAD1, and HUS1. (A) Left: Visualization of the FISH signal for RAD17. Right: DAPI stain of the metaphase spread to identify chromosome 5. (B) Visualization of the FISH signal for RAD1 and DAPI stain of the mitotic figure to identify chromosomes 5 and 10. (C) Visualization of the FISH signal for HUS1 and DAPI stain of the mitotic figure to identify chromosome 7. FIG. 6. Results of simultaneous FISH mapping for human RAD17 and RAD1. Left: Visualization of the FISH signals for RAD17 and RAD1. Right: DAPI stain of the metaphase spread to identify chromosome 5. man homolog of MEC1 and rad3, while ATM is the in S. cerevisiae. Although the checkpoint control sys- homolog of S. cerevisiae TEL1 and another, recently tems of S. cerevisiae and S. pombe have several gene identified, S. pombe gene (Bentley et al., 1996; Cim- homologs in common, they appear to have diverged prich et al., 1996; Carr, 1997; Bentley and Carr, 1997). significantly, perhaps because cell division is different The human rad checkpoint gene complement ap- in these organisms; S. cerevisiae cells divide by bud- pears to match more closely the set of genes present in ding while S. pombe divides by fission. The mitosis and S. pombe, compared to S. cerevisiae. One of the S. cell division of S. pombe is more similar to that of cerevisiae checkpoint genes, MEC3, does not appear to human cells than the S. cerevisiae process is; unlike S. have S. pombe or human homologs, while homologs for cerevisiae, S. pombe has a distinct G2 phase of the cell S. pombe hus1 and rad26 have not yet been identified cycle, and in addition the chromosomes undergo con- HUMAN RAD CHECKPOINT HOMOLOGS 433

TABLE 2 the location of an unidentified tumor suppressor gene. Rad Checkpoint Gene Homologs Allelic deletion mapping identified a deletion desig- nated del-27 at 5p13–p12 as being associated with Human small cell lung carcinomas (Wieland et al., 1996). Tu- chromosome mor-specific loss of heterozygosity was also detected at S. cerevisiae S. pombe Human map location del-27 in 10 of 38 (26%) bladder cancers (Bohm et al., 1997). Loss of 5p13 is also associated with colon cancer RAD24 rad17 RAD17a 5q13 RAD17 rad1 RAD1 5p14–p13.2 development (Yeatman et al., 1996). In addition, dele- MEC3 —— tion of 5p13 was identified as causing developmental RAD9 rhp9 — defects including microcephaly, hypertonia, microgna- — hus1 HUS1 7p13–p12 thia, and mental retardation (Keppen et al., 1992). This DDC1 rad9 RAD9 11q13 — rad26 — is reminiscent of the pleiotropic phenotype of ataxia MEC1 rad3 ATR 3q24–q22 telangiectasia, a multisystem disorder involving TEL1 Yes ATM 11q22 ataxia, due to neurodegeneration; telangiectases (patches of dilated blood vessels) in the face, eyes, and a Similar gene names do not necessarily indicate homology. For ears; and immunodeficiencies. As a result, RAD1 may example, S. pombe rad17 and S. cerevisiae RAD17 are not homologs of each other nor are S. pombe rad9 and S. cereviviae RAD9. be a candidate for the tumor suppressor gene that resides at 5p13. Chromosome rearrangements at 7p13, especially densation during mitosis (Russell and Nurse, 1986; translocations, have been noted in non-Hodgkin lym- Carr and Hoekstra, 1995). phoma (Jonveaux et al., 1990; Dyer et al., 1993) and Presence of Candidate Tumor Suppressor Genes at acute lymphoblastic leukemia (Uckun et al., 1987, 1989). In addition, these translocations occur fre- Human Checkpoint Gene Chromosome Loci quently in the lymphocytes of patients with ataxia- A number of human cell cycle regulatory genes have telangiectasia, Nijmegen breakage syndrome, or re- been previously identified as tumor suppressors, and lated disorders characterized by heightened rates of the human homologs of the S. pombe rad checkpoint chromosome breakage. However, in these cases it is genes are also potential candidates for a role in tumor felt that the observed chromosome rearrangements are suppression. A distinction has been made between two an effect, not a cause, of the syndromes (Conley et al., types of cancer-susceptibility genes, termed “gatekeep- 1986; Hecht and Hecht, 1987; Taalman et al., 1989; ers” and “caretakers” (Kinzler and Vogelstein, 1997). Barbi et al., 1991; Green et al., 1995; Renedo et al., Genes that control cell growth directly are gatekeep- 1997). The chromosome rearrangements occur almost ers, while genes that monitor and maintain the integ- exclusively in lymphocytes, primarily at just four sites: rity of the genome are caretakers. Inactivation of care- 7p13, 7q34, 14q11, and 14q32. These four regions act taker genes leads to tumor formation indirectly, by as though they contain fragile sites limited to lympho- increasing the overall mutation rate and the likelihood cytes. The fragile sites may be the T-cell receptor ␣, ␤, of inactivation of a gatekeeper gene. If the RAD check- and ␥ chain and immunoglobulin H genes that map to point genes turn out to act as tumor suppressors they these sites and undergo recombination during lympho- will most likely be of the caretaker class of cancer- cyte development. Errors in recombination at these loci susceptibility genes. probably explain the chromosome 7 and 14 rearrange- Many studies have previously identified human ments. Indeed, these rearrangements are not un- chromosome 5q13 as the location of an unidentified common in normal lymphocytes, occurring with a tumor suppressor gene. Translocations, deletions, and frequency of 5 ϫ 10Ϫ4 per metaphase in phytohemag- inversions involving 5q13 have been found to be asso- glutinin-stimulated lymphocyte cultures. The conse- ciated with hematologic malignancies (Fairman et al., quences of most of these events appear to be benign or 1996; Shanske et al., 1996; Gogineni et al., 1996; Mor- of little clinical significance (Dewald et al., 1986; gan et al., 1988). Rearrangements of band 5q13 are also Scheres et al., 1986; Hecht et al., 1987). Due to the high associated with chondrosarcomas (Ordnal et al., 1993; background of benign chromosomal rearrangements at Tarkkanen et al., 1993). Losses of 5q13–q21 (Miura et 7p13 in lymphocytes, it appears difficult to determine al., 1992) and rearrangement involving 5q13–q31 whether potential tumor suppressors for hematologic (Goguel et al., 1995) have been observed in small cell malignancies lie within this region. However, the re- lung cancers. A loss of heterozygosity was found at gion 7p13–p22 has been identified as being rearranged 5q13.1–q21 in ovarian cancer (Tavassoli et al., 1996), in a number of cases of ovarian cancer (Pejovic, 1995), and translocation involving 5q13 is a candidate for and so HUS1 may be a candidate tumor suppressor for primary chromosome changes in renal cancer (Berger ovarian carcinogenesis. et al., 1986). RAD17 would appear to be a candidate for Mapping of the human chromosomal location for the tumor suppressor gene that has been identified in these three putative cell cycle regulatory genes con- the 5q13 region. firms their attractiveness as candidate tumor suppres- Human chromosome 5p13 has also been identified as sor genes. Future work will be important in determin- 434 DEAN, LIAN, AND O’DONNELL ing whether these genes actually do play such a role in currence, breakpoints, and clinical and biological significance. humans. Am. J. Hum. Genet. 38: 520–532. Dyer, M. J. S., Nacheva, E., Fischer, P., Heward, J. M., Labastide, W., and Karpas, A. (1993). A new human T-cell lymphoma cell line ACKNOWLEDGMENTS (Karpas 384) of the T-cell receptor gamma/delta lineage with translocation t(7:14)(p13;q11.2). Leukemia 7: 1047–1053. We thank Dr. Yuji Kohara, National Institute of Genetics, Japan; Enoch, T., and Nurse, P. (1990). Mutation of fission yeast cell cycle Dr Hee-Sup Shin, Pohang Institute of Science & Technology, Repub- control genes abolishes dependence of mitosis on DNA replication. lic of Korea; and Dr. Graeme Bell, The University of Chicago, for gifts Cell 60: 665–673. of clones. We thank the Protein/DNA Technology Center of The Enoch, T., Carr, A. M., and Nurse, P. (1992). Fission yeast genes Rockefeller University for copious amounts of DNA sequence deter- involved in coupling mitosis to completion of DNA replication. mination. This work was supported by NIH Grant GM54705. Genes Dev. 6: 2035–2046. Enoch, T., and Norbury, C. (1995). Cellular responses to DNA dam- age: Cell-cycle checkpoints, apoptosis and the roles of p53 and REFERENCES ATM. Trends Biochem. Sci. 20: 426–430. Fairman, J., Wang, R. Y., Liang, H., Zhao, L., Saltman, D., Liang, Al-Khodairy, F., and Carr, A. M. (1992). DNA repair mutants defin- J. C., and Nagarajan, L. (1996). Translocations and deletions of ing G2 checkpoint pathways in Schizosaccharomyces pombe. 5q13.1 in myelodysplasia and acute myelogenous leukemia: Evi- EMBO J. 11: 1343–1350. dence for a novel critical locus. Blood 88: 2259–2266. Al-Khodairy, F., Fotou, E., Sheldrick, K. S., Griffiths, D. J. F., Leh- Gogineni, S. K., da Costa, M., and Verma, M. S. (1996). A new mann, A. R., and Carr, A. M. (1994). Identification and character- translocation, t(5;21)(q13;q22) in acute myelogenous leukemia. ization of new elements involved in checkpoint and feedback con- Cancer Genet. Cytogenet. 88: 167–169. trols in fission yeast. Mol. Biol. Cell 5: 147–160. Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J. Goguel, A. F., Fouquet, F., Duverger, A., Arvelo, F., Jacrot, M., (1990). Basic local alignment search tool. J. Mol. Biol. 215: 403– Poupon, M. F., and Bernheim, A. (1995). Evolution of chromosomal 410. alterations and biologic features in two small cell lung carcinoma cell lines established from one patient during the course of the Altschul, S. F., Madden, T. L., Scha¨ffer, A. A., Zhang, J., Zhang, Z., disease. Cancer Genet. Cytogenet. 80: 47–54. Miller, W., and Lipman, D. J. (1997). Gapped BLAST and PSI- BLAST: A new generation of protein database search programs. Green, A. J., Yates, J. R. W., Taylor, A. M. R., Biggs, P., McGuire, Nucleic Acids Res. 25: 3389–3402. G. M., McConville, C. M., Billing, C. J., and Barnes, N. D. (1995). Barbi, G., Scheres, J. M. J. C., Schindler, D., Taalman, R. D. F. M., Severe microcephaly with normal intellectual development: The Rodens, K., Mehnert, K., Mueller, M., and Seyschab, H. (1991). Nijmegem breakage syndrome. Arch. Dis. Child. 73: 431–434. Chromosome instability and X-ray hypersensitivity in a microce- Griffiths, D. J. F., Barbet, N. C., McReady, S., Lehmann, A. R., and phalic and growth-retarded child. Am. J. Med. Genet. 40: 44–50. Carr, A. M. (1995). Fission yeast rad17: A homolog of budding Bentley, N. J., and Carr, A. M. (1997). DNA structure-dependent yeast RAD24 that shares regions of sequence similarity with DNA checkpoints in model systems. Biol. Chem. 378: 1267–1274. polymerase accessory proteins. EMBO J. 14: 5812–5823. Bentley, N. J., Holtzman, D. A., Flaggs, G., Keegan, S., DeMaggio, Gyapay, G., Schmitt, K., Fizames, C., Jones, H., Vega-Czarny, N., A., Ford, J. C., Hoekstra, M., and Carr, A. M. (1996). The Schizo- Spillett, D., Muselet, D., Prud’Homme, J. F., Dib, C., Auffray, C., saccharomyces pombe rad3 checkpoint gene. EMBO J. 15: 6641– Morissette, J., Weissenbach, J., and Goodfellow, P. N. (1996). A 6651. radiation hybrid map of the . Hum. Mol. Genet. 5: Berger, C. S., Sandberg, A. A., Todd, I. A., Pennington, R. D., 339–346. Haddad, F. S., Hecht, B. K., and Hecht, F. (1986). Chromosomes in Hartwell, L. (1992). Defects in a cell cycle checkpoint may be respon- kidney, ureter, and bladder cancer. Cancer Genet. Cytogenet. 23: sible for the genomic instability of cancer cells. Cell 71: 543–546. 1–24. Hartwell, L., and Kastan, M. B. (1994). Cell cycle control and cancer. Bohm, M., Kirch, H., Otto, T., Rubben, H., and Wieland, I. (1997). Science 266: 1821–1828. Deletion analysis at the DEL-27, APC and MTS1 loci in bladder Hartwell, L., and Weinert, T. A. (1989). Checkpoints: Controls that cancer: LOH at the DEL-27 locus on 5p13–12 is a prognostic ensure the order of cell cycle events. Science 246: 629–634. marker of tumour progression. Int. J. Cancer 74: 291–295. Hecht, F., and Hecht, B. K. (1987). Chromosome changes connect sc/Rad3sp Carr, A. M. (1997). Control of cell cycle arrest by the Mec1 immunodeficiency and cancer in ataxia–telangiectasia. Am. J. Pe- DNA structure checkpoint pathway. Curr. Opin. Genet. Dev. 7: diatr. Hematol. Oncol. 9: 185–188. 93–98. Hecht, F., Hecht, B. K., and Kirsch, I. R. (1987). Fragile sites limited Carr, A. M., and Hoekstra, M. F. (1995). The cellular responses to to lymphocytes: Molecular recombination and malignancy. Cancer DNA damage. Trends Cell Biol. 5: 32–40. Genet. Cytogenet. 26: 95–104. Cimprich, K. A., Shin, T. B., Keith, C. T., and Schreiber, S. L. (1996). cDNA cloning and gene mapping of a candidate human cell cycle Heng, H. H. Q., Squire, J., and Tsui, L. C. (1992). High resolution checkpoint protein. Proc. Natl. Acad. Sci. USA 93: 2850–2855. mapping of mammalian genes by in situ hybridization to free chromatin. Proc. Natl. Acad. Sci. USA 89: 9509–9513. Collins, K., Jacks, T., and Pavletich, N. P. (1997). The cell cycle and cancer. Proc. Natl. Acad. Sci. USA 94: 2776–2778. Heng, H. H. Q., and Tsui, L. C. (1993). Modes of DAPI banding and simultaneous in situ hybridization. Chromosoma 102: 325–332. Conley, M. E., Spinner, N. B., Emanuel, B. S., Nowell, P. C., and Nichols, W. W. (1986). Chromosomal breakage syndrome with Holden, D. W., Spanos, A., and Banks, G. R. (1989). Nucleotide profound immunodeficiency. Blood 67: 1251–1256. sequence of the REC1 gene of Ustilago maydis. Nucleic Acids Res. Cox, D. R., Burmeister, M., Price, E. R., Kim, S., and Myers, R. M. 17: 10489. (1990). Radiation hybrid mapping: A somatic cell genetic method Holliday, R., Halliwell, R. E., Evans, M. W., and Rowell, V. (1976). for constructing high-resolution maps of mammalian chromo- Genetic characterization of rec-1, a mutant of Ustilago maydis somes. Science 250: 245–250. defective in repair and recombination. Genet. Res. 27: 413–453. Dewald, G. W., Noonan, K. J., Spurbeck, J. L., and Johnson, D. D. Jackson, S. P. (1996). The recognition of DNA damage. Curr. Opin. (1986). T-lymphocytes with 7;14 translocations: Frequency of oc- Genet. Dev. 6: 19–25. HUMAN RAD CHECKPOINT HOMOLOGS 435

Jimenez, G., Yucel, J., Rowley, R., and Subramani, S. (1992). The Dissection of DNA replication and G2 checkpoint control function. rad3ϩ gene of Schizosaccharomyces pombe is involved in multiple Mol. Gen. Genet. 255: 332–340. checkpoint functions and in DNA repair. Proc. Natl. Acad. Sci. Miura, I., Graziano, S. L., Cheng, J. Q., Doyle, L. A., and Testa, J. R. USA 89: 4952–4956. (1992). Chromosome alterations in human small cell lung cancer: Jonveaux, P., Le Coniat, M., Derre, J., Vecchione, D., and Berger, R. Frequent involvement of 5q. Cancer Res. 52: 1322–1328. (1990). Deletion of (7p13p14) in non-Hodgkin’s lymphoma. Cancer Morgan, R., Walter, T. A., Decker, H. J., Hecht, F., and Sandberg, Genet. Cytogenet. 50: 53–6. A. A. (1988). Inversion of chromosome 5 long arm in region of cell Kanter-Smoler, G., Knudsen, K. E., Jimenez, G., Sunnerhagen, P., growth gene cluster in hematologic disorders. Cancer Genet. Cyto- and Subramani, S. (1995). Separation of phenotypes in mutant genet. 32: 267–275. alleles of the Schizosaccharomyces pombe cell-cycle checkpoint ϩ Murray, J. M., Carr, A. M., Lehmann, A. R., and Watts, F. Z. (1991). gene rad1 . Mol. Biol. Cell 6: 1793–1805. Cloning and characterization of the rad9 DNA repair gene from Kastan, M. B., Onyekwere, O., Sidransky, D., Vogelstein, B., and Schizosaccharomyces pombe. Nucleic Acids Res. 19: 3525–3531. Craig, R. W. (1991). Participation of p53 protein in the cellular Nasim, A., and Smith, B. P. (1975). Genetic control of radiation response to DNA damage. Cancer Res. 51: 6304–6311. sensitivity in Schizosaccharomyces pombe. Genetics 79: 573–582. Kelman, Z., and O’Donnell, M. (1995). DNA polymerase III holoen- Onel, K., Thelen, M. P., Ferguson, D. O., Bennett, R. L., and Hollo- zyme: Structure and function of a chromosomal replicating ma- man, W. K. (1995). Mutation avoidance and DNA repair profi- chine. Annu. Rev. Biochem. 64: 171–200. ciency in Ustilago maydis are differentially lost with progressive Keppen, L. D., Gollin, S. M., Edwards, D., Sawyer, J., Wilson, W., truncation of the REC1 gene product. Mol. Cell. Biol. 15: 5329– and Overhauser, J. (1992). Clinical phenotype and molecular anal- 5338. ysis of a three-generation family with an interstitial deletion of the Onel, K., Koff, A., Bennett, R. L., Unrau, P., and Holloman, W. K. short arm of chromosome 5. Am. J. Med. Genet. 44: 356–360. (1996). The REC1 gene of Ustilago maydis, which encodes a 3Ј–5Ј Kinzler, K. W., and Vogelstein, B. (1997). Gatekeepers and caretak- exonuclease, couples DNA repair and completion of DNA synthesis ers. Nature 386: 761–762. to a mitotic checkpoint. Genetics 143: 165–174. Koonin, E. V. (1993). A common set of conserved motifs in a vast Ordnal, C., Mandahl, N., Rydholm, A., Willen, H., Brosjo, O., and variety of putative nucleic acid-dependent ATPases including Mitelman, F. (1993). Chromosome aberrations and cytogenetic MCM proteins involved in the initiation of eukaryotic DNA repli- intratumor heterogeneity in chondrosarcomas. J. Cancer Res. cation. Nucleic Acids Res. 21: 2541–2547. Clin. Oncol. 120: 51–56. Koonin, E. V. (1997). A conserved ancient domain joins the growing Parker, A. E., Van de Weyer, I., Laus, M. C., Oostveen, I., Yon, J., superfamily of 3Ј–5Ј exonucleases. Curr. Biol. 7: R604–606. Verhasselt, P., and Luyten, W. H. M. L. (1998a). A human homo- Kostrub, C. F., Al-Khodairy, F., Ghazizadeh, H., Carr, A. M., and logue of the Schizosaccharomyces pombe rad1ϩ checkpoint gene ϩ Enoch, T. (1997). Molecular analysis of hus1 , a fission yeast gene encodes an exonuclease. J. Biol. Chem. 273: 18332–18339. required for S-M and DNA damage checkpoints. Mol. Gen. Genet. Parker, A. E., Van de Weyer, I., Laus, M. C., Verhasselt, P., and 254: 389–399. Luyten, W. H. M. L. (1998b). Identification of a human homologue Kostrub, C. F., Knudsen, K., Subramani, S., and Enoch, T. (1998). of the Schizosaccharomyces pombe rad17ϩ checkpoint gene. Hus1p, a conserved fission yeast checkpoint protein, interacts with J. Biol. Chem. 273: 18340–18346. Rad1p and is phosphorylated in response to DNA damage. EMBO J. 7: 2055–2066. Pejovic, T. (1995). Genetic changes in ovarian cancer. Ann. Med. 27: 73–78. Kuerbitz, S. J., Plunkett, B. S., Walsh, W. V., and Kastan, M. B. (1992). Wild-type p53 is a cell cycle checkpoint determinant fol- Renedo, M., Robledo, M., Arranz, E., Infantes, F., Roman, A., Garcia- lowing irradiation. Proc. Natl. Acad. Sci. USA 89: 7491–7495. Yebenes, J., and Benitez, J. (1997). Cytogenetic and molecular studies of siblings with ataxia telangiectasia followed for 7 years. Lehmann, A. R., and Carr, A. M. (1995). The ataxia–telangiectasia Cancer Genet. Cytogenet. 95: 178–182. gene: A link between checkpoint controls, neurodegeneration, and cancer. Trends Genet. 11: 375–377. Rowley, R., Subramani, S., and Young, P. G. (1992). Checkpoint controls in Schizosaccharomyces pombe: rad1. EMBO J. 11: 1335– Lieberman, H. B., Hopkins, K. M., Nass, M., Demetrick, D., and 1342. Davey, S. (1996). A human homolog of the Schizosaccharomyces pombe rad9ϩ checkpoint control gene. Proc. Natl. Acad. Sci. USA Russell, P., and Nurse, P. (1986). Schizosaccharomyces pombe and 93: 13890–13895. Saccharomyces cerevisiae: A look at yeasts divided. Cell 45: 781– Long, K. E., Sunnerhagen, P., and Subramani, S. (1994). The Schizo- 782. saccharomyces pombe rad1 gene consists of three exons and the Savitsky, K., Bar-Shira, A., Gilad, S., Rotman, G., Ziv, Y., Vanagaite, cDNA sequence is partially homologous to the Ustilago maydis L., Tagle, D. A., Smith, S., Uziel, T., Sfez, Ashkenazi, M., Pecker, REC1 cDNA. Gene 148: 155–159. I., Frydman, M., Harnik, R., Patanjali, S. R., Simmons, A., Clines, Longhese, M. P., Paciotti, V., Fraschini, R., Zaccarini, R., Plevani, P., G. A., Sartiel, A., Gatti, R. R., Chessa, L., Sanal, O., Lavin, M. F., and Lucchini, G. (1997). The novel DNA damage checkpoint pro- Jaspers, N. G. J., Taylor, A. M. R., Arlett, C. F., Miki, T., Weiss- tein Ddc1p is phosphorylated periodically during the cell cycle and man, S. M., Lovett, M., Collins, F. S., and Shiloh, Y. (1995). A in response to DNA damage in budding yeast. EMBO J. 16: 5216– single ataxia telangiectasia gene with a product similar to PI-3 5226. kinase. Science 268: 1749–1753. Lydall, D., and Weinert, T. (1995). Yeast checkpoint genes in DNA Scheres, J. M., Hustinx, T. W., and Trent, J. M. (1986). Possible damage processing: Implications for repair and arrest. Science involvement of unstable sites on chromosomes 7 and 14 in human 270: 1488–1491. cancer. Cancer Genet. Cytogenet. 19: 151–158. Lydall, D., and Weinert, T. (1996). From DNA damage to cell cycle Shanske, A. L., Kalman, A., and Grunwald, H. (1996). A myelopro- arrest and suicide: A budding yeast perspective. Curr. Opin. Genet. liferative disorder with eosinophila associated with a unique Dev. 6: 4–11. translocation (3;5). Br. J. Haematol. 95: 524–526. Lydall, D., and Weinert, T. (1997). G2/M checkpoint genes of Sac- Sheldrick, K. S., and Carr, A. M. (1993). Feedback controls and G2 charomyces cerevisiae: Further evidence for roles in DNA replica- checkpoints: Fission yeast as a model system. BioEssays 15: 775– tion and/or repair. Mol. Gen. Genet. 256: 638–651. 782. McFarlane, R. J., Carr, A. M., and Price, C. (1997). Characterization Siede, W., Nusspaumer, G., Portillo, V., Rodriguez, R., and Fried- of the Schizosaccharomyces pombe rad4/cut5 mutant phenotypes: berg, E. C. (1996). Cloning and characterization of RAD17, a gene 436 DEAN, LIAN, AND O’DONNELL

controlling cell cycle responses to DNA damage in Saccharomyces Uckun, F. M., Kersey, J. H., Gajl-Peczalska, K. J., Heerema, N. A., cerevisiae. Nucleic Acids Res. 24: 1669–1675. Provisor, A. J., Haag, D., Gilchrist, G., Song, C. W., Arthur, D. C., Stewart, E., and Enoch, T. (1996). S-phase and DNA-damage check- Roloff, J., Lampkin, B., Greenwood, M., Dewald, G., and Vallera, points: A tale of two yeasts. Curr. Opin. Cell Biol. 8: 781–787. D. A. (1987). Heterogeneity of cultured leukemic lymphoid progen- Sunnerhagen, P., Seaton, B. L., Nasim, A., and Subramani, S. (1990). itor cells from B cell precursor acute lymphoblastic leukemia Cloning and analysis of a gene involved in DNA repair and recom- (ALL) patients. J. Clin. Invest. 80: 639–646. bination, the rad1 gene of Schizosaccharomyces pombe. Mol. Cell. Uckun, F. M., Gajl-Peczalska, K. J., Provisor, A. J., and Heerema, Biol. 10: 3750–3760. N. A. (1989). Immunophenotype–karyotype associations in human Taalman, R. D. F. M., Hustinx, T. W. J., Weemaes, C. M. R., See- acute lymphoblastic leukemia. Blood 73: 271–280. manova, E., Schmidt, A., Passarge, E., and Scheres, M. J. C. Weinert, T. A., and Hartwell, L. H. (1990). Characterization of RAD9 (1989). Further delineation of the Nijmegen breakage syndrome. of Saccharomyces cerevisiae and evidence that its function acts Am. J. Med. Genet. 32: 425–431. posttranslationally in cell cycle arrest after DNA damage. Mol. Tarkkanen, M., Kaipainen, A., Karaharju, E., Bohling, T., Szyman- Cell. Biol. 10: 6554–6564. ska, J., Helio, H., Kivioja, A., Elomaa, I., and Knuutila, S. (1993). Weinert, T. A., Kiser, G. L., and Hartwell, L. H. (1994). Mitotic Cytogenetic study of 249 consecutive patients examined for a bone checkpoint genes in budding yeast and the dependence of mitosis tumor. Cancer Genet. Cytogenet. 68: 1–21. on DNA replication and repair. Genes Dev. 8: 652–665. Tavassoli, M., Steingrimsdottir, H., Pierce, E., Jiang, X., Alagoz, M., Wieland, I., Bohm, M., Arden, K. C., Ammermuller, T., Bogatz, S., Farzaneh, F., and Campbell, I. G. (1996). Loss of heterozygosity on Viars, C. S., and Rajewsky, M. F. (1996). Allelic deletion mapping chromosome 5q in ovarian cancer is frequently accompanied by on chromosome 5 in human carcinomas. Oncogene 12: 97–102. TP53 mutation and identifies a tumour suppressor gene locus at Willson, J., Wilson, S., Warr, N., and Watts, F. Z. (1997). Isolation 5q13.1–21. Br. J. Cancer 74: 115–119. and characterization of the Schizosaccharomyces pombe rhp9 Thelen, M. P., Onel. K., and Holloman, W. K. (1994). The REC1 gene gene: A gene required for the DNA damage checkpoint but not the of Ustilago maydis involved in the cellular response to DNA dam- replication checkpoint. Nucleic Acids Res. 25: 2138–2146. age encodes an exonuclease. J. Biol. Chem. 269: 747–754. Yeatman, T. J., Cher, M. L., Mao, W., Wloch, M., and Tedesco, T. Tsukuda, T., Bauchwitz, R., and Holloman, W. K. (1989). Isolation of (1996). Identification of genetic alterations associated with the the REC1 gene controlling recombination in Ustilago maydis. Gene process of human experimental colon cancer liver metastasis in the 85: 335–341. nude mouse. Clin. Exp. Metastasis 14: 246–252.