- Email: [email protected]
RecA-Related Gene
GENOMICS
49, 103–111 (1998) GE985226
ARTICLE NO.
Identification, Characterization, and Genetic Mapping of Rad51d, a New Mouse and Human RAD51/RecA-Related Gene Douglas L. Pittman, Leah Rosa Weinberg, and John C. Schimenti1 The Jackson Laboratory, 600 Main Street, Bar Harbor, Maine 04609 Received November 24, 1997; accepted January 21, 1998
Homologous DNA recombination occurs in all organisms and is important for repair of DNA damage during mitosis. One of the critical genes for DNA repair and meiotic recombination in yeast is RAD51, and homologs of RAD51 have been identified in several species, including mouse and human. Here we describe a new RAD51-related mammalian gene, named Rad51d, identified by searching the EST database with the yeast RAD55 and human RAD51B/REC2 genes. A fulllength 1.5-kb mouse cDNA clone that encodes a predicted 329-amino-acid protein was isolated. Rad51d mRNA was present in every mouse tissue examined. Four different transcript sizes were detected, one of which was specific to testis. Human cDNA clones that predicted 71% amino acid identity to the mouse protein were also isolated. Interestingly, the sequences of these human clones and of RT-PCR-derived products provided evidence for alternative splicing. These mRNAs are predicted to encode proteins that are truncated relative to the mouse and lack the ATP-binding motif characteristic of RecA-related proteins. Using an interspecific backcross mapping panel, Rad51d was mapped to mouse Chromosome 11, 48.5 cM from the centromere. By radiation hybrid mapping, the human ortholog RAD51D was mapped to chromosome 17q11, which is a region syntenic to mouse Chromosome 11. Due to its expression pattern and sequence similarity to other RAD51 family members, it is likely that Rad51d is part of a complex of proteins required for DNA repair and meiotic recombination. q 1998 Academic Press
INTRODUCTION
Homologous recombination is critical for species survival. In somatic cells, recombination is required for maintenance of genome stability and for VDJ recombination of immunoglobulin and T-cell receptor genes. Sequence data from this article have been deposited with the EMBL/GenBank Data Libraries under Accession Nos. AF034955 (Rad51d) and AF034956 (RAD51D). 1 To whom correspondence should be addressed. Telephone: (207) 288-6402. Fax: (207) 288-6082. E-mail: [email protected].
Defects in mitotic recombination result in the inability to repair DNA damage properly, and mutations in a number of DNA repair genes have been associated with human diseases, including PMS1, PMS2, MSH2, and MLH1 (Nicolaides et al., 1994; Fishel et al., 1993; Bronner et al., 1994). In germ cells, meiotic recombination is critical for proper pairing and segregation of homologous chromosomes. Defects in meiotic recombination result in unequal chromosome distribution, potentially causing infertility or birth defects, as in Downs syndrome (Lamb et al., 1996). Analysis of mutations affecting recombination in fungi has been crucial in understanding recombination mechanics and has served to guide experimental design in mammalian studies. In Saccharomyces cerevisiae, the Rad51 protein has been shown to be related to the Escherichia coli RecA protein, showing the strongest similarity in a domain that interacts with the ATP cofactor. The RecA protein coats single-stranded DNA, forming a helical filament, and promotes pairing and strand transfer between homologous DNA molecules in an ATP-dependent manner (for reviews, see Radding, 1991, and West, 1992). Rad51 forms filaments on double-stranded DNA and also promotes strand exchange (Ogawa et al., 1993; Sung, 1994). The RAD55, RAD57, and DMC1 yeast genes encode proteins that also share sequence homology with RecA in the ATP cofactorbinding region. All but one of these genes (DMC1) were originally identified as genes required for DNA repair (Game and Mortimer, 1974). One interpretation of this observation is that the mitotic DNA repair genes have been recruited for meiotic recombination (Game, 1993). Genetic and biochemical evidence suggests that these RecA-like proteins form parts of a complex. Affinity chromatography studies and two-hybrid analysis demonstrated that the Rad52 and Rad51 proteins interact and that this association requires only one-half of the N-terminal region of Rad51 and one-third of the C-terminal portion of Rad52 (Shinohara et al., 1992; Milne and Weaver, 1993; Donovan et al., 1994). Twohybrid analysis has also demonstrated that Rad51 associates with itself, with Rad55, and with Rad54 and that Rad55 interacts with Rad57 (Johnson and Symington, 1995; Hays et al., 1995; Clever et al., 1997).
103
AID
GENO 5226
/
6r64$$$301
03-13-98 05:33:31
0888-7543/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
gnma
104
PITTMAN, WEINBERG, AND SCHIMENTI
Consistent with these results, RAD51 and RAD52 on a high-copy plasmid suppressed the X-ray sensitivity of rad55 and rad57 mutants (Hays et al., 1995; Johnson and Symington, 1995). DMC1 is a meiosis-specific member of the RAD51 family identified in yeast (Bishop et al., 1992). Mutations in this gene block meiotic progression, synaptonemal complex formation, and homologous recombination. Immunolocalization studies have demonstrated that the Dmc1 protein is bound to approximately 64 sites along meiotic chromosomes (Bishop, 1994), colocalized with the Rad51 protein, and required Rad51 for chromosome binding. These results support the idea that the Rad51, Rad52, Rad54, Rad55, Rad57, and Dmc1 proteins interact to form protein complexes required for recombination and DNA repair. A number of the genes critical to DNA repair and meiotic recombination in yeast have homologs in mammals. RAD51 homologs in mouse and human were isolated based upon their identity to the S. cerevisiae and Schizosaccharomyces pombe RAD51 genes (Morita et al., 1993; Shinohara et al., 1993). A mouse Rad51 cDNA (now called Rad51a) was found to complement the S. cerevisiae rad51 mutation, and the human homolog was shown to have ATP-dependent strand transfer activity (Morita et al., 1993; Baumann et al., 1996; Gupta et al., 1997). The mouse Rad51a gene is expressed at high levels in ovary and testis and is associated with the axial/lateral element in synaptonemal complexes in mouse spermatocytes and oocytes (Shinohara et al., 1993; Haaf et al., 1995; Ashley et al., 1995; Plug et al., 1996). The Rad51 protein is detected early in meiosis as small, evenly dispersed foci (270 in spermatocytes, 350 in oocytes), and by the end of leptotene, 32–38 larger foci are detected in both sexes, suggesting that the protein complexes recognized by a Rad51 antibody form even larger complexes during chromosomal condensation (Plug et al., 1996). Targeted mutagenesis of the mouse Rad51a gene resulted in early embryonic lethality (Tsuzuki et al., 1996), in contrast to the viable yeast phenotype of rad51 mutants. Mammalian DMC1 homologs were isolated due to their sequence similarity to DMC1 homologs in S. cerevisiae, S. pombe, and Lilium (Habu et al., 1996; Sato et al., 1995a,b). The mouse and human genes encode 340-amino-acid proteins containing the nucleotide-binding motif region similar to RecA. Transcription of the mouse Dmc1 gene is testisspecific, consistent with its proposed role in meiotic recombination, but expression in human is detected in a variety of tissues, suggesting that Dmc1 may have a role in somatic cells in some mammalian species. Recently, a new RAD51-like mammalian homolog, RAD51B/REC2 (Rice et al., 1997; Albala et al., 1998), was identified by its similarity to yeast RAD51 and REC2 in the fungus Ustilago maydis. RAD51B/REC2 was expressed in every human tissue examined, and its expression was induced by radiation, consistent with its proposed role in DNA repair. We have identified yet another member of the RAD51 gene family, named Rad51d, discovered by expressed
AID
GENO 5226
/
6r64$$$302
03-13-98 05:33:31
sequence tag (EST) database searches with Rad51 family members (Adams et al., 1995; Lennon et al., 1996). The predicted proteins share similarity with other RAD51-like proteins, suggesting that they have a role in recombination and DNA repair. Here we describe the initial characterization and chromosomal mapping of the mouse and human Rad51d genes. MATERIALS AND METHODS Clone identification. To identify new potential mouse and human homologs of the RAD51-gene family, the S. cerevisiae RAD55 (Lovett, 1994) and human REC2/RAD51B (Rice et al., 1997; Albala et al., 1998) amino acid sequences were used to search the NCBI EST database on the XREF server (Bassett et al., 1995, 1997) and the TIGR Human Gene Index. Two new overlapping mouse cDNAs [I.M.A.G.E. Consortium (LLNL) cDNA clones 746795 and 478233] and two new overlapping human cDNAs were identified [I.M.A.G.E. Consortium (LLNL) cDNA Clones 60163 and 259579]. The 746795 clone was generated from mouse whole fetus, 12.5 dpc, and the 478233 clone was identified from 13.5- to 14.5-dpc whole mouse fetus. The 60163 clone was from human adult T-lymphocytes, and the 259579 clone was from 8- to 9-week-pc human placenta. The cDNA clones were obtained from either American Type Culture Collection (Rockville, MD), or Research Genetics (Huntsville, AL). The 259579, 746795, and 478233 cDNA clones were in the EcoRI and NotI sites of the pT7T3D-Pac vector (Pharmacia) (Lennon et al., 1996). The 60163 clone was in the EcoRI and XhoI sites of the pBluescript SK0 vector (Stratagene). Sequencing, informatics, and nomenclature. Sequencing was performed on a 373 automated sequencer (ABI) with Taq Dye-Deoxy Terminators (ABI), and all clones were completely sequenced on both strands. To confirm the nucleotide sequence of the human RAD51D gene, the oligonucleotide primer pair RAD51D-5* [(5*-CGAGGAGATGATCCAGCTTCTCA-3*) corresponding to /39 bp (see Fig. 1B)] and RAD51D-3* [(5*-CTCCAGTATAGAGACCAGCATCAA-3*) corresponding to /184 bp] was used to amplify DNA extracted from a human neuronal cDNA library constructed from a retinoic acid-induced NT2 cell line (Ackerman et al., 1994) as follows. Following a 2-min denaturing step at 977C, 35 rounds of amplification were performed: 947C for 30 s, 607C for 30 s, and 727C for 1 min. Both strands of the resulting amplimer were sequenced using the same primers. The nucleotide sequence was assembled using the Sequencer 3.0 software (Gene Codes Corp., Ann Arbor, MI). Amino acid and nucleotide comparisons were performed using the GeneWorks 2.4N sequence analysis program (IntelliGenetics; Mountain View, CA), and gaps in alignments were counted as one mismatch. Database searches and sequence comparisons were performed at NCBI (USA) and GenomeNet (Japan). pI and molecular weights were calculated using the pI/MW tool available on the ExPasY molecular biology server (Bjellqvist et al., 1994). This is the fourth RAD51-like gene identified in human based upon sequence similarity to the yeast RAD51 gene. In keeping with the naming system established by the Human Nomenclature Committee, this new gene has been given the name RAD51D in human and Rad51d in mouse. The gene sequences were assigned Accession Nos. AF034955 (Rad51d) and AF034956 (RAD51D). Northern blot analysis. A mouse multiple tissue Northern (MTN) blot (Clontech 7762-1) was hybridized with a mouse Rad51d probe (897-bp EcoRI/SacI fragment from clone 478233) for 1 h at 657C in ExpressHyb (Clontech). Two room-temperature washes in 11 SSC (150 mM NaCl, 15 mM sodium citrate, pH 7), 0.1% SDS were performed for 10 min, followed by one wash at 507C in 0.11 SSC, 0.1% SDS for 20 min. The filter was exposed to a phosphorimaging screen and analyzed on a Fuji BAS 2000 analysis system. The filter was stripped in 0.5% SDS (90–1007C) for 20 min. The human b-actin gene (Clontech) was then used to probe the same filter. Hybridization and wash conditions were the same as described above, except the two final washes were at 657C in 0.11 SSC, 0.1% SDS for 20 min.
gnma
IDENTIFICATION OF Rad51d Zoo blot. A Southern blot filter (Clontech 7753-1), containing 4 mg of EcoRI-digested genomic DNA from human, monkey, rat, mouse, dog, cow, rabbit, chicken, and yeast was hybridized at 657C in Church buffer (Church and Gilbert, 1984) with an 897-bp EcoRI–SacI fragment from the 478233 mouse cDNA clone. Two washes were performed in 11 SSC, 0.1% SDS at room temperature for 10 min and once at 657C 11 SSC, 0.1% SDS for 10 min. Signal was detected on a Fuji BAS 2000 system as described above. Genetic mapping of Rad51d in mouse and human. Mouse Rad51d was mapped on the BSS (C57BL/6J 1 Mus spretus) 1 M. spretus interspecific backcross mapping panel maintained at The Jackson Laboratory as follows. A restriction site polymorphism between the parental strains was identified in a 457-bp DNA fragment generated by PCR amplification at the 3* end of Rad51d. The C57BL/6J allele, but not the M. spretus allele, was cleavable by MspI to yield fragment sizes of 377 and 80 bp. The 5* and 3* primers were 478233.4L (5*GCCCAGAATTACCTGGCAAG-3*) and 478UTR.1241 (5*-GCAGATGGGAAAACAGACCA-3*), respectively. Following a 3-min denaturing step at 947C, 35 rounds of amplification were performed: 947C for 45 s, 557C for 1 min, and 727C for 2 min. The amplification products were then digested with MspI prior to agarose gel electrophoresis. Mapping data and further information on the backcross panel can be found on the World Wide Web at http://www.jax.org/resources/ documents/cmdata/. To map human RAD51D, oligonucleotide primer pair 60163.PCR1 (5*-CTGGCGACTGGATGGATAATC-3*) and 60163.PCR2 (5*-CCCAGTAACTCAGAGACAGAG-3*) was used. Following a 2-min denaturing step at 977C, 35 rounds of amplification were performed: 947C for 30 s, 557C for 30 s, 727C for 1 min. No signal was detected from hamster genomic DNA using these amplification conditions. A 185bp amplimer from the 3* end of RAD51D (1179–1364) was generated from human genomic DNA, and a panel of hamster–human hybrids was screened (Stanford Radiation Hybrid mapping panel G3, Research Genetics). Data were submitted to the Stanford Radiation Hybrid Server ([email protected]), which returned the linkage data reported under Results.
RESULTS
Identification of a New RAD51-Related Gene in Mouse and Human As part of an effort to identify new mammalian genes required for recombination, the amino acid sequences of S. cerevisiae RAD55 (Lovett, 1994) and human RAD51B/REC2 (Rice et al., 1997; Albala et al., 1997) were used to search the NCBI EST database, using the XREF Server (Bassett et al., 1995, 1996, 1997) and the TIGR Human Gene Index. An EST, I.M.A.G.E. Consortium clone ID 478233 (GenBank Accession No. AA049653), was identified, having a match of P value 2.7 1 1003 to RAD55 and 1.1 1 1004 to human RAD51B/REC2. The complete DNA sequence of this clone, which originated from a 13.5- to 14.5-day mouse embryo cDNA library, was determined and used to search the EST division of GenBank. An additional I.M.A.G.E. clone extending 404 bp further 5* was identified from a 12.5-day mouse embryonic library (clone ID 746795; GenBank Accession No. AA260430). Identical matches to this new mouse gene were I.M.A.G.E. clones 692411 (GenBank Accession No. AA237779) and 849350 (GenBank Accession No. AA457827), originating from mouse liver and mammary gland cDNA libraries, respectively. Due to its significant similarity to other RAD51 family members, this new mouse gene has been named Rad51d (see Materials and Methods).
AID
GENO 5226
/
6r64$$$302
03-13-98 05:33:31
105
Sequence analysis suggests that the 746795 mouse cDNA clone is full length (Fig. 1). Translation of a 987bp ORF predicts a 329-amino-acid protein with a molecular mass of 35,260 Da and a pI of 6.76. The sequence surrounding the presumed start codon has optimal context for a translation initiation site, an A at 03 and G at /4, and in-frame stop codons are present upstream of the presumed start codon (Kozak, 1996). No AATAAA polyadenylation signals were present at the 3* end, but a poly(A) tail was present on both Rad51d EST clones. Further GenBank queries with the mouse Rad51d nucleotide sequence revealed two highly significant hits with human ESTs, I.M.A.G.E. Clone 60163 (GenBank Accession No. AA352205) from human T-lymphocytes and Clone 259579 (GenBank Accession No. N57184) from human placenta. Both appear to be full length in the sense that they contain the initiation codon. Overall, the two clones showed extensive sequence similarity to mouse Rad51d with similar predicted start and stop sites (Fig. 1). However, each has unusual structural differences. The shorter of the two cDNA clones, 259579, appears to represent an alternatively spliced product; it excludes sequences corresponding to two exons in the mouse gene (D. Pittman, unpublished observation), one encoding the first ATP-binding motif (see Fig. 1 legend), and encodes a 216-amino-acid protein. The 60163 EST clone also contains an apparent alternative splice that skips a single exon present in the mouse gene (D. Pittman, unpublished observation; see Fig. 1). The absence of this exon results in the introduction of a premature stop codon due to a frameshift, thereby predicting a truncated protein containing only 49 amino acids. To exclude the possibility that the frameshift in 60163 was not a cloning artifact, a PCR product encompassing these nucleotides was generated from a human neuronal cDNA library (see Materials and Methods). The nucleotide sequence was identical to the sequence of the 60163 EST clone. If the frameshift is ignored (Fig. 1B), the 60163 insert would encode a 289-amino-acid protein sharing 71% overall amino acid identity and 73% overall nucleotide identity with the mouse gene, suggesting that the human and mouse genes are orthologs (see also syntenic mapping data below). Accordingly, the human gene has been assigned the name RAD51D (see Materials and Methods). The theoretical molecular mass of the protein encoded by the larger human cDNA clone, 60163, is 30,960 Da with a pI of 5.87. Similarity of Rad51d/RAD51D to the RAD51 Family of Genes Searches of the databases using the S. cerevisiae Rad51 protein produced an extremely large number of significant hits with other RAD51 homologs, obscuring more weakly related genes. Even though the original Rad51d clone, 478233, was identified in a search against RAD55, Rad51d shows the highest level of
gnma
106
PITTMAN, WEINBERG, AND SCHIMENTI
amino acid similarity to RAD51 genes present in several organisms (e.g., yeast, chick, mouse, and human). This relatedness is illustrated in Fig. 1C by alignment with the mouse Rad51a member of the RAD51 gene family. Like the other members of the RAD51 gene family, mouse and human Rad51d/RAD51D genes have two conserved ATP-binding domains (Fig. 1). The first (GGPGSGKTQ) is located between amino acids 107 and 115 in the mouse sequence and between amino acids 67 and 75 in the human sequence. Another highly conserved region is a second ATP-binding domain (VVVIVDS) present at amino acids 202–207 in the Rad51d sequence and at amino acids 163–168 in the human RAD51D protein. Finally, the residues NH at positions 249–250 in Rad51d (and also in RAD51D), which are conserved as either NH or NQ in other RecArelated genes and are thought to be involved in ATPinduced conformational change, are present in these genes (Story et al., 1993). Conservation of Rad51d To determine the degree of conservation of Rad51d in other mammalian species, a Southern blot of genomic DNA from seven mammalian species was hybridized using Rad51d (an 897-bp EcoRI/SacI fragment from the 478233 clone) as the probe (Fig. 2). Strong signals were detected in DNA from all mammalian species tested and from yeast (S. cerevisiae). The signal detected from yeast genomic DNA is most likely RAD51 due to its similarity to Rad51d (19% identity at the amino acid level). The size of the fragment is also consistent with the size predicted from the genomic sequence around RAD51, approximately 6.1 kb. A weak signal was also detected in chicken. These results indicated conservation of the Rad51d gene across mammalian species. Rad51d Tissue Distribution To determine the size and tissue distribution of the Rad51d transcripts in mouse, a Northern blot, containing poly(A)/ RNA from several mouse tissues, was hybridized with a 32P-labeled Rad51d probe (Fig. 3). Multiple transcripts, 7.7, 2.5, and 1.7 kb, were detected in every tissue examined. The smaller transcript is consistent with the size of the mouse cDNA clones. The 7.7-kb transcript appears to be the strongest signal in brain tissue, and the two smaller transcripts appear to be strongest in all other tissues. A 3.6-kb testis-specific transcript was also detected. Rad51d Maps to Mouse Chromosome 11 and RAD51D Maps to Human Chromosome 17 Using a restriction site difference between parental strains of The Jackson Laboratory’s interspecific mouse
backcross (see Materials and Methods), Rad51d was mapped to position 48.5 (according to the Mouse Genome Database) on Chromosome 11 (lod score 25.9), in an interval that also contains the microsatellite marker D11Mit36 (Fig. 4). A scan of the region for mapped traits that might be consistent with a mutation in Rad51d revealed the pulmonary adenoma resistance gene (Par1). Par1 is a quantitative trait locus (QTL) in M. spretus that confers partial resistance to lung tumor susceptibility present in the strain A/J (Manenti et al., 1996). The chromosomal location of human RAD51D was identified using the Stanford Radiation Hybrid mapping panel G3. The human RAD51D gene was mapped within 7.7 cR from STS WI-2760 (aka, D17S822E, EST00213) with a lod score of 13.6 (Fig. 5). The WI2760 marker is located at 58–65 cM on chromosome 17, placing RAD51D at 17q11 within a 7-cM region flanked by markers D17S933 and D17S791. These regions of mouse Chromosome 11 and human chromosome 17 share synteny, suggesting that these two genes are direct orthologs. DISCUSSION
Recombinational Repair and the RecA-Related Genes Systematic identification of eukaryotic genes involved in recombinational repair has been conducted predominantly in the yeast S. cerevesiae. Of the numerous experimental advantages afforded by this organism, the ability to screen haploid cells for recessive mutations was particularly important in the isolation of repair mutants. Because ionizing radiation causes DNA double-stranded breaks, and double-stranded breaks are repaired by a recombinational mechanism, several mutations isolated in screens for sensitivity to X rays ultimately proved to affect genes involved in recombination. RAD51 was one of several yeast genes identified in screens for X-ray-sensitive mutants (Game and Mortimer, 1974; Game, 1993). RAD51 is related to the E. coli RecA protein, which is involved in homologous recombination and DNA repair (see Edelmann and Kucherlapati, 1996). Indeed, rad51 mutants are defective in mitotic recombination, indicative of Rad51’s role in a recombination repair system. RAD55 and RAD57 are two additional yeast genes required for efficient repair of ionizing radiation-induced damage that share homology to RecA, and rad55 and rad57 mutants are also recombination-defective (Game and Mortimer, 1974; Lovett and Mortimer, 1987). Several lines of evidence point to the existence of
FIG. 1. Nucleotide and predicted amino acid sequences of (A) mouse Rad51d and (B) human RAD51D. The two conserved nucleotide binding domains are underlined. The position of an apparent frameshift, due to the possible absence of a single exon, in human I.M.A.G.E. clone 60163 is indicated by NN at nucleotide position 146 (see text). Note that the I.M.A.G.E. clone 259579 does not contain basepairs 145 to 363. (C) Amino acid alignment of the mouse and human Rad51d proteins along with the amino acid sequence of mouse Rad51a. (Dots represent identical amino acids; dashes indicate gaps in the amino acid alignment.)
AID
GENO 5226
/
6r64$$$303
03-13-98 05:33:31
gnma
IDENTIFICATION OF Rad51d
AID
GENO 5226
/
6r64$$5226
03-13-98 05:33:31
gnma
107
108
PITTMAN, WEINBERG, AND SCHIMENTI
sis-specific). In addition to having orthologs of RAD51 and DMC1, at least three other homologs have been discovered in mice (and humans): RAD51B/Rec2, RAD51C, and Rad51d/RAD51D. It is not clear what the orthologs of these last three genes are, whether they are direct relatives of RAD51 itself or of DMC1, RAD55, and RAD57. The situation may be clarified as additional RecA homologs are identified or as the actual biochemical functions of the genes are elucidated. A truncated form of RAD51D in humans was identified from cDNAs generated from two different tissues. A second splice variant that lacked the nucleotide-binding motif critical for the function of RAD51 was identified. These alternatively spliced variants raise some interesting questions as to whether the peptide made from the latter transcript is functional and whether posttranscriptional regulation is occurring at the level of RNA processing. A mutation in the ATP-binding residue in the yeast RAD57 gene did not lead to loss of activity (Johnson and Symington, 1995). Perhaps this motif is not critical for the function of RAD51D or lack of this motif is complemented by another RAD51 family member present in a recombinational complex of proteins. Isolation of additional RAD51D cDNAs and the FIG. 2. Conservation of Rad51d. An 897-bp EcoRI/SacI fragment from the mouse 478233 cDNA clone was used to probe a Southern blot containing approximately 4 mg of EcoRI-digested genomic DNA from several mammalian species, chicken, and yeast. Size markers are indicated on the left.
a recombinational repair system in mammals. First, human and mouse homologs of fungal recombinational repair genes have recently been identified. These include RAD51A (Shinohara et al., 1993; Morita et al., 1993), RAD52 (Bendixen, 1994; Muris et al., 1994; Shen et al., 1995), Rec2/RAD51B (Rice et al., 1997; Albala et al., 1997), which is a RAD51 homolog, RAD51C (a Rad51-related gene isolated in the laboratory of D. Schild, pers. comm., Berkeley, CA, October 1997), and RAD51D. Second, as with yeast RAD51, strand exchange activity has been demonstrated for the human RAD51 protein, indicating conserved function (Baumann et al., 1996; Ogawa et al., 1993; Sung, 1994). Third, site-specific double-strand breaks induced in mammalian cells by a rare-cutting endonuclease can be repaired by homologous recombination in addition to end-joining (Rouet et al., 1994; Sargent et al., 1997). However, the majority of stimulated repair events are nonhomologous, indicating that repair of double-strand breaks by recombination is not the major pathway in mitotic cells. Fourth, mammalian homologs of the yeast RAD genes have been shown to interact (see Golub et al., 1997). It appears that mammals may have an expanded RecA family, relative to yeast. Because the sequence of the S. cerevesiae genome has been determined, we know that yeast contains four RecA-related genes: RAD51, RAD55, RAD57 and DMC1 (the latter is meio-
AID
GENO 5226
/
6r64$$$304
03-13-98 05:33:31
FIG. 3. Expression analysis of mouse Rad51d. (Top) A mouse tissue blot was probed with an 897-bp EcoRI/SacI fragment from the I.M.A.G.E. clone 478233. The Rad51d transcript sizes are indicated on the right. (Bottom) A b-actin cDNA probe was used to verify the presence of RNA in each lane.
gnma
IDENTIFICATION OF Rad51d
109
Whether Rad51d is involved in either of these genetic disorders requires further study. Recombinational Repair Genes and Meiosis Genetic control of recombination in mammals has been exceedingly difficult to study, because tools do not exist to screen efficiently for mutations that may result in meiotic failure (hence, sterility). However, investigations in yeast have shown that several proteins crucial for meiotic recombination are also involved in mitotic recombination and vice versa. Mutations in RAD51, RAD55, and RAD57 affect meiotic as well as mitotic recombination (Game, 1993). It is reasonable to assume that the same will hold true for mammalian recombinational repair genes. Therefore, in the pursuit of meiotic recombination genes, a potentially fruitful, more technically feasible approach may be to identify genes involved in mitotic recombinational repair. One obvious avenue is to screen the EST database for homologs of known yeast recombination genes, as we have accomplished here. There is also a practical advantage to this strategy: large-scale EST sequencing efforts have yet to conduct extensive sequencing of germ-cell cDNA libraries. Furthermore, although clones from entire re-
FIG. 4. Localization of Rad51d on mouse Chromosome 11. The gene order and relative marker positions are shown. Map distance in centimorgans is shown on the left.
genomic sequence will help resolve these questions. Furthermore, despite the overall similarity to Rad51a, it is not certain whether Rad51d encodes similar biochemical function. For example, yeast RAD55 and RAD57 are not directly involved in strand exchange, but serve as accessory proteins to stabilize or enhance the activity of RAD51 (Sung, 1997). Should Rad51d be involved in DNA repair, then mutations in this gene might cause phenotypes such as tumor susceptibility or genome instability. Inspection of the mouse Chromosome 11 genetic map near Rad51d revealed close linkage to a QTL for pulmonary adenoma resistance (Par1) (Manenti et al., 1996). The A/J strain of mice has a susceptible allele at this locus, whereas the M. spretus allele confers resistance in heterozygotes. Furthermore, the human homolog (RAD51D) maps to 17q11.2, near the FWT1 gene that confers Wilms tumor susceptibility (Rahman et al., 1996). This gene has been mapped in an interval (17q12–q21) that is apparently just distal to RAD51D.
AID
GENO 5226
/
6r64$$$304
03-13-98 05:33:31
FIG. 5. Chromosomal localization of human RAD51D. Idiogram of human chromosome 17 showing linkage of RAD51D to D17S822E as determined by radiation hybrid mapping using the G3 Stanford Radiation Hybrid Panel (1 cR equals approximately 30 kb). Distances, in cR, between outside markers, and lod scores are presented. Map distances were taken from the Stanford RH Mapping Index at http://www-shgc.stanford.edu/Mapping/rh/.
gnma
110
PITTMAN, WEINBERG, AND SCHIMENTI
productive organs are being sequenced, meiotic prophase staged cells represent only a subset of all cell types. Hence, genes expressed at relatively low levels at specific meiotic stages may escape sequence determination. Nevertheless, RAD51-related genes in mammals have been identified using degenerate PCR approaches, as in the case of REC2/RAD51B, RAD51, and DMC1 (Rice et al., 1997; Morita et al., 1993; Shinohara et al., 1993; Sato et al., 1995a). Current evidence suggests roles for mammalian RAD51A in meiotic recombination. Mouse Rad51a is expressed at high levels in ovary and testis, and the protein is associated with the axial/lateral element in synaptonemal complexes in mouse spermatocytes and oocytes (Shinohara et al., 1993; Ashley et al., 1995; Haaf et al., 1995; Plug et al., 1996). Unfortunately, targeted mutagenesis of Rad51a resulted in early embryonic lethality, precluding a study of its possible role in meiosis (Tsuzuki et al., 1996). It is possible that similar problems may be encountered with targeted mutants of the other RAD51-related recombinational repair genes. In light of this, it appears that conditional mutations may be required to understand the in vivo roles of these genes in meiotic vs mitotic cells. Additionally, the possible effects of mutations in these genes on recombination may require assays, possibly employing transgenes or PCR, capable of measuring gene conversion and crossing over in situations where infertility may result. ACKNOWLEDGMENTS The authors thank Mary Barter and Lucy Rowe for assistance in mapping Rad51d using the BSS backcross, Doug McMinimy and Amy Lambert for automated sequencing, Patsy Nishina for providing DNA from the recombination hybrid mapping panel, Joyce Worcester for preparing figures, David Schild and Larry Thompson for providing unpublished information, Sue Ackerman for providing the human cDNA library, and Patsy Nishina and Tim O’Brien for critically reading the manuscript. This work was supported by NIH Grant GM45415 to J.S. and by a Cancer Center core grant (CA34196) to The Jackson Laboratory. D.P. was supported by NIH Training Grant HD07065 from the National Institute of Child Health and Human Development, by a fellowship from the Lalor Foundation, and by NRSA Award GM19052.
REFERENCES Ackerman, S. L., Knowles, B. B., and Andrews, P. W. (1994). Gene regulation during neuronal and non-neuronal differentiation of NTERA2 human teratocarcinoma-derived stem cells. Brain Res. Mol. Brain Res. 25: 157–162. Adams, M. D., Kerlavage, A. R., Fleischmann, R. D., Fuldner, R. A., Bult, C. J., Lee, N. H., Kirkness, E. F., Weinstock, K. G., Gocayne, J. D., White, O., et al. (1995). Initial assessment of human gene diversity and expression patterns based upon 83 million nucleotides of cDNA sequence. Nature 377: 3–174. Albala, J. S., Thelen, M. P., Prange, C., Fan, W., Christensen, M., Thompson, L. H., and Lennon, G. G. (1998). Identification of a novel human RAD51 homolog, RAD51B. Genomics, in press. Ashley, T., Plug, A. W., Xu, J., Solari, A. J., Reddy, G., Golub, E. I., and Ward, D. C. (1995). Dynamic changes in Rad51 distribution
AID
GENO 5226
/
6r64$$$304
03-13-98 05:33:31
on chromatin during meiosis in male and female vertebrates. Chromosoma 104: 19–28. Bassett, D. E., Jr., Boguski, M. S., and Hieter, P. (1996). Yeast genes and human disease. Nature 379: 589–590. [Letter] Bassett, D. E., Jr., Boguski, M. S., Spencer, F., Reeves, R., Goebl, M., and Hieter, P. (1995). Comparative genomics, genome crossreferencing and XREFdb. Trends Genet. 11: 372–373. Bassett, D. E., Jr., Boguski, M. S., Spencer, F., Reeves, R., Kim, S., Weaver, T., and Hieter, P. (1997). Genome cross-referencing and XREFdb: Implications for the identification and analysis of genes mutated in human disease. Nat. Genet. 15: 339–344. Baumann, P., Benson, F. E., and West, S. C. (1996). Human Rad51 protein promotes ATP-dependent homologous pairing and strand transfer reactions in vitro. Cell 87: 757–766. Bendixen, C., Sunjevaric, I., Bauchwitz, R., and Rothstein, R. (1994). Identification of a mouse homologue of the Saccharomyces cerevisiae recombination and repair gene, RAD52. Genomics 23: 300– 303. Bishop, D. K. (1994). RecA homologs Dmc1 and Rad51 interact to form multiple nuclear complexes prior to meiotic chromosome synapsis. Cell 79: 1081–1092. Bishop, D. K., Park, D., Xu, L., and Kleckner, N. (1992). DMC1: A meiosis-specific yeast homolog of E. coli recA required for recombination, synaptonemal complex formation, and cell cycle progression. Cell 69: 439–456. Bjellqvist, B., Basse, B., Olsen, E., and Celis, J. E. (1994). Reference points for comparisons of two-dimensional maps of proteins from different human cell types defined in a pH scale where isoelectric points correlate with polypeptide compositions. Electrophoresis 15: 529–539. Bronner, C. E., Baker, S. M., Morrison, P. T., Warren, G., Smith, L. G., Lescoe, M. K., Kane, M., Earabino, C., Lipford, J., Lindblom, A., et al. (1994). Mutation in the DNA mismatch repair gene homologue hMLH1 is associated with hereditary non-polyposis colon cancer. Nature 368: 258–261. Church, G., and Gilbert, W. (1984). Genomic sequencing. Proc. Natl. Acad. Sci. USA 81: 1991–1995. Clever, B. I. H., Schmuckli-Maurer, J., King, J., Sigrist, M., and Heyer, W. D. (1997). Recombinational repair in yeast: Functional interactions between Rad51 and Rad54 proteins. EMBO J. 16: 2535–2544. Donovan, J. W., Milne, G. T., and Weaver, D. T. (1994). Homotypic and heterotypic protein associations control Rad51 function in double-strand break repair. Genes Dev. 8: 2552–2562. Edelmann, W., and Kucherlapati, R. (1996). Role of recombination enzymes in mammalian cell survival. Proc. Natl. Acad. Sci. USA 93: 6225–6227. Fishel, R., Lescoe, M. K., Rao, M. R., Copeland, N. G., Jenkins, N. A., Garber, J., Kane, M., and Kolodner, R. (1993). The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer. Cell 75: 1027–1038. Game, J. C. (1993). DNA double-strand breaks and the RAD50RAD57 genes in Saccharomyces. Cancer Biol. 4: 73–83. Game, J. C., and Mortimer, R. K. (1974). A genetic study of X-ray sensitive mutants in yeast. Mutat. Res. 24: 281–292. Golub, E. I., Kovalenko, O. V., Gupta, R. C., Ward, D. C., and Radding, C. M. (1997). Interaction of human recombination proteins Rad51 and Rad54. Nucleic Acids Res. 25: 4106–4110. Gupta, R. C., Bazemore, L. R., Golub, E. I., and Radding, C. M. (1997). Activities of human recombination protein Rad51. Proc. Natl. Acad. Sci. USA 94: 463–468. Haaf, T., Golub, E. I., Reddy, G., Radding, C. M., and Ward, D. C. (1995). Nuclear foci of mammalian Rad51 recombination protein in somatic cells after DNA damage and its localization in synaptonemal complexes. Proc. Natl. Acad. Sci. USA 92: 2298–2302. Habu, T., Taki, T., West, A., Nishimune, Y., and Morita, T. (1996). The mouse and human homologs of DMC1, the yeast meiosis-spe-
gnma
IDENTIFICATION OF Rad51d cific homologous recombination gene, have a common unique form of exon-skipped transcript in meiosis. Nucleic Acids Res. 24: 470– 477. Hays, S. L., Firmenich, A. A., and Berg, P. (1995). Complex formation in yeast double-strand break repair: Participation of Rad51, Rad52, Rad55, and Rad57 proteins. Proc. Natl. Acad. Sci. USA 92: 6925–6929. Johnson, R. D., and Symington, L. S. (1995). Functional differences and interactions among the putative RecA homologs Rad51, Rad55, and Rad57. Mol. Cell. Biol. 15: 4843–4850. Kozak, M. (1996). Interpreting cDNA sequences: Some insights from studies on translation. Mamm. Genome 7: 563–574. Lamb, N. E., Freeman, S. B., Savage Austin, A., Pettay, D., Taft, L., Hersey, J., Gu, Y., Shen, J., Saker, D., May, K. M., Avramopoulos, D., Petersen, M. B., Hallberg, A., Mikkelsen, M., Hassold, T. J., and Sherman, S. L. (1996). Susceptible chiasmate configurations of chromosome 21 predispose to non-disjunction in both maternal meiosis I and meiosis II. Nat. Genet. 14: 400–405. Lennon, G., Auffray, C., Polymeropoulos, M., and Soares, M. B. (1996). The I.M.A.G.E. consortium: An integrated molecular analysis of genomes and their expression. Genomics 33: 151–152. Lovett, S. T. (1994). Sequence of the RAD55 gene of Saccharomyces cerevisiae: Similarity of RAD55 to prokaryotic RecA and other RecA-like proteins. Gene 142: 103–106. Lovett, S. T., and Mortimer, R. K. (1987). Characterization of null mutants of the RAD55 gene of Saccharomyces cerevisiae: Effects of temperature, osmotic strength and mating type. Genetics 116: 547–553. Manenti, G., Gariboldi, M., Elango, R., Fiorino, A., De Gregorio, L., Falvella, F. S., Hunter, K., Housman, D., Pierotti, M. A., and Dragani, T. A. (1996). Genetic mapping of a pulmonary adenoma resistance (Par1) in mouse. Nat. Genet. 12: 455–457. Milne, G. T., and Weaver, D. T. (1993). Dominant negative alleles of RAD52 reveal a DNA repair/recombination complex including Rad51 and Rad52. Genes Dev. 7: 1755–1765. Morita, T., Yoshimura, Y., Yamamoto, A., Murata, K., Mori, M., Yamamoto, H., and Matsushiro, A. (1993). A mouse homolog of the Escherichia coli recA and Saccharomyces cerevisiae RAD51 genes. Proc. Natl. Acad. Sci. USA 90: 6577–6580. Muris, D. F., Bezzubova, O., Buerstedde, J. M., Vreeken, K., Balajee, A. S., Osgood, C. J., Troelstra, C., Hoeijmakers, J. H., Ostermann, K., Schmidt, H., et al. (1994). Cloning of human and mouse genes homologous to RAD52, a yeast gene involved in DNA repair and recombination. Mutat. Res. 315: 295–305. Nicolaides, N. C., Papadopoulos, N., Liu, B., Wei, Y. F., Carter, K. C., Ruben, S. M., Rosen, C. A., Haseltine, W. A., Fleischmann, R. D., Fraser, C. M., et al. (1994). Mutations of two PMS homologues in hereditary nonpolyposis colon cancer. Nature 371: 75–80. Ogawa, T., Yu, X., Shinohara, A., and Egelman, E. H. (1993). Similarity of the yeast RAD51 filament to the bacterial RecA filament. Science 259: 1896–1899. Plug, A. W., Xu, J., Reddy, G., Golub, E. I., and Ashley, T. (1996).
AID
GENO 5226
/
6r64$$$305
03-13-98 05:33:31
111
Presynaptic association of Rad51 protein with selected sites in meiotic chromatin. Proc. Natl. Acad. Sci. USA 93: 5920–5924. Radding, C. M. (1991). Helical interactions in homologous pairing and strand exchange driven by RecA protein. J. Biol. Chem. 266: 5355–5358. Rahman, N., Arbour, L., Tonin, P., Renshaw, J., Pelletier, J., Baruchel, S., Pritchard Jones, K., Stratton, M. R., and Narod, S. A. (1996). Evidence for a familial Wilms’ tumour gene (FWT1) on chromosome 17q12–q21. Nat. Genet. 13: 461–463. Rice, M. C., Smith, S. T., Bullrich, F., Havre, P., and Kmiec, E. B. (1997). Isolation of human and mouse genes based on homology to REC2, a recombinational repair gene from the fungus Ustilago maydis. Proc. Natl. Acad. Sci. USA 94: 7417–7422. Rouet, P., Smih, F., and Jasin, M. (1994). Expression of a site-specific endonuclease stimulates homologous recombination in mammalian cells. Proc. Natl. Acad. Sci. USA 91: 6064–6068. Sargent, R. G., Brenneman, M. A., and Wilson, J. H. (1997). Repair of site-specific double-strand breaks in a mammalian chromosome by homologous and illegitimate recombination. Mol. Cell. Biol. 17: 267–277. Sato, S., Kobayashi, T., Hotta, Y., and Tabata, S. (1995a). Characterization of a mouse recA-like gene specifically expressed in testis. DNA Res. 2: 147–150. Sato, S., Seki, N., Hotta, Y., and Tabata, S. (1995b). Expression profiles of a human gene identified as a structural homologue of meiosis-specific recA-like genes. DNA Res. 2: 183–186. Shen, Z., Denison, K., Lobb, R., Gatewood, J. M., and Chen, D. J. (1995). The human and mouse homologs of the yeast RAD52 gene: cDNA cloning, sequence analysis, assignment to human chromosome 12p12.2–p13, and mRNA expression in mouse tissues. Genomics 25: 199–206. Shinohara, A., Ogawa, H., Matsuda, Y., Ushio, N., Ikeo, K., and Ogawa, T. (1993). Cloning of human, mouse and fission yeast recombination genes homologous to RAD51 and recA. Nat. Genet. 4: 239–243. Shinohara, A., Ogawa, H., and Ogawa, T. (1992). Rad51 protein involved in repair and recombination in S. cerevisiae is a RecA-like protein. Cell 69: 457–470. Story, R. M., Bishop, D. K., Kleckner, N., and Steitz, T. A. (1993). Structural relationship of bacterial RecA proteins to recombination proteins from bacteriophage T4 and yeast. Science 259: 1892– 1896. Sung, P. (1994). Catalysis of ATP-dependent pairing and strand exchange by yeast RAD51 protein. Science 265: 1241–1243. Sung, P. (1997). Yeast Rad55 and Rad57 proteins form a heterodimer that functions with replication protein A to promote DNA strand exchange by Rad51 recombinase. Genes Dev. 11: 1111–1121. Tsuzuki, T., Fujii, Y., Sakumi, K., Tominaga, Y., Nakao, K., Sekiguchi, M., Matsushiro, A., Yoshimura, Y., and Morita, T. (1996). Targeted disruption of the Rad51 gene leads to lethality in embryonic mice. Proc. Natl. Acad. Sci. USA 93: 6236–6240. West, S. C. (1992). Enzymes and molecular mechanisms of genetic recombination. Annu. Rev. Biochem. 61: 603–640.
gnma
49, 103–111 (1998) GE985226
ARTICLE NO.
Identification, Characterization, and Genetic Mapping of Rad51d, a New Mouse and Human RAD51/RecA-Related Gene Douglas L. Pittman, Leah Rosa Weinberg, and John C. Schimenti1 The Jackson Laboratory, 600 Main Street, Bar Harbor, Maine 04609 Received November 24, 1997; accepted January 21, 1998
Homologous DNA recombination occurs in all organisms and is important for repair of DNA damage during mitosis. One of the critical genes for DNA repair and meiotic recombination in yeast is RAD51, and homologs of RAD51 have been identified in several species, including mouse and human. Here we describe a new RAD51-related mammalian gene, named Rad51d, identified by searching the EST database with the yeast RAD55 and human RAD51B/REC2 genes. A fulllength 1.5-kb mouse cDNA clone that encodes a predicted 329-amino-acid protein was isolated. Rad51d mRNA was present in every mouse tissue examined. Four different transcript sizes were detected, one of which was specific to testis. Human cDNA clones that predicted 71% amino acid identity to the mouse protein were also isolated. Interestingly, the sequences of these human clones and of RT-PCR-derived products provided evidence for alternative splicing. These mRNAs are predicted to encode proteins that are truncated relative to the mouse and lack the ATP-binding motif characteristic of RecA-related proteins. Using an interspecific backcross mapping panel, Rad51d was mapped to mouse Chromosome 11, 48.5 cM from the centromere. By radiation hybrid mapping, the human ortholog RAD51D was mapped to chromosome 17q11, which is a region syntenic to mouse Chromosome 11. Due to its expression pattern and sequence similarity to other RAD51 family members, it is likely that Rad51d is part of a complex of proteins required for DNA repair and meiotic recombination. q 1998 Academic Press
INTRODUCTION
Homologous recombination is critical for species survival. In somatic cells, recombination is required for maintenance of genome stability and for VDJ recombination of immunoglobulin and T-cell receptor genes. Sequence data from this article have been deposited with the EMBL/GenBank Data Libraries under Accession Nos. AF034955 (Rad51d) and AF034956 (RAD51D). 1 To whom correspondence should be addressed. Telephone: (207) 288-6402. Fax: (207) 288-6082. E-mail: [email protected].
Defects in mitotic recombination result in the inability to repair DNA damage properly, and mutations in a number of DNA repair genes have been associated with human diseases, including PMS1, PMS2, MSH2, and MLH1 (Nicolaides et al., 1994; Fishel et al., 1993; Bronner et al., 1994). In germ cells, meiotic recombination is critical for proper pairing and segregation of homologous chromosomes. Defects in meiotic recombination result in unequal chromosome distribution, potentially causing infertility or birth defects, as in Downs syndrome (Lamb et al., 1996). Analysis of mutations affecting recombination in fungi has been crucial in understanding recombination mechanics and has served to guide experimental design in mammalian studies. In Saccharomyces cerevisiae, the Rad51 protein has been shown to be related to the Escherichia coli RecA protein, showing the strongest similarity in a domain that interacts with the ATP cofactor. The RecA protein coats single-stranded DNA, forming a helical filament, and promotes pairing and strand transfer between homologous DNA molecules in an ATP-dependent manner (for reviews, see Radding, 1991, and West, 1992). Rad51 forms filaments on double-stranded DNA and also promotes strand exchange (Ogawa et al., 1993; Sung, 1994). The RAD55, RAD57, and DMC1 yeast genes encode proteins that also share sequence homology with RecA in the ATP cofactorbinding region. All but one of these genes (DMC1) were originally identified as genes required for DNA repair (Game and Mortimer, 1974). One interpretation of this observation is that the mitotic DNA repair genes have been recruited for meiotic recombination (Game, 1993). Genetic and biochemical evidence suggests that these RecA-like proteins form parts of a complex. Affinity chromatography studies and two-hybrid analysis demonstrated that the Rad52 and Rad51 proteins interact and that this association requires only one-half of the N-terminal region of Rad51 and one-third of the C-terminal portion of Rad52 (Shinohara et al., 1992; Milne and Weaver, 1993; Donovan et al., 1994). Twohybrid analysis has also demonstrated that Rad51 associates with itself, with Rad55, and with Rad54 and that Rad55 interacts with Rad57 (Johnson and Symington, 1995; Hays et al., 1995; Clever et al., 1997).
103
AID
GENO 5226
/
6r64$$$301
03-13-98 05:33:31
0888-7543/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
gnma
104
PITTMAN, WEINBERG, AND SCHIMENTI
Consistent with these results, RAD51 and RAD52 on a high-copy plasmid suppressed the X-ray sensitivity of rad55 and rad57 mutants (Hays et al., 1995; Johnson and Symington, 1995). DMC1 is a meiosis-specific member of the RAD51 family identified in yeast (Bishop et al., 1992). Mutations in this gene block meiotic progression, synaptonemal complex formation, and homologous recombination. Immunolocalization studies have demonstrated that the Dmc1 protein is bound to approximately 64 sites along meiotic chromosomes (Bishop, 1994), colocalized with the Rad51 protein, and required Rad51 for chromosome binding. These results support the idea that the Rad51, Rad52, Rad54, Rad55, Rad57, and Dmc1 proteins interact to form protein complexes required for recombination and DNA repair. A number of the genes critical to DNA repair and meiotic recombination in yeast have homologs in mammals. RAD51 homologs in mouse and human were isolated based upon their identity to the S. cerevisiae and Schizosaccharomyces pombe RAD51 genes (Morita et al., 1993; Shinohara et al., 1993). A mouse Rad51 cDNA (now called Rad51a) was found to complement the S. cerevisiae rad51 mutation, and the human homolog was shown to have ATP-dependent strand transfer activity (Morita et al., 1993; Baumann et al., 1996; Gupta et al., 1997). The mouse Rad51a gene is expressed at high levels in ovary and testis and is associated with the axial/lateral element in synaptonemal complexes in mouse spermatocytes and oocytes (Shinohara et al., 1993; Haaf et al., 1995; Ashley et al., 1995; Plug et al., 1996). The Rad51 protein is detected early in meiosis as small, evenly dispersed foci (270 in spermatocytes, 350 in oocytes), and by the end of leptotene, 32–38 larger foci are detected in both sexes, suggesting that the protein complexes recognized by a Rad51 antibody form even larger complexes during chromosomal condensation (Plug et al., 1996). Targeted mutagenesis of the mouse Rad51a gene resulted in early embryonic lethality (Tsuzuki et al., 1996), in contrast to the viable yeast phenotype of rad51 mutants. Mammalian DMC1 homologs were isolated due to their sequence similarity to DMC1 homologs in S. cerevisiae, S. pombe, and Lilium (Habu et al., 1996; Sato et al., 1995a,b). The mouse and human genes encode 340-amino-acid proteins containing the nucleotide-binding motif region similar to RecA. Transcription of the mouse Dmc1 gene is testisspecific, consistent with its proposed role in meiotic recombination, but expression in human is detected in a variety of tissues, suggesting that Dmc1 may have a role in somatic cells in some mammalian species. Recently, a new RAD51-like mammalian homolog, RAD51B/REC2 (Rice et al., 1997; Albala et al., 1998), was identified by its similarity to yeast RAD51 and REC2 in the fungus Ustilago maydis. RAD51B/REC2 was expressed in every human tissue examined, and its expression was induced by radiation, consistent with its proposed role in DNA repair. We have identified yet another member of the RAD51 gene family, named Rad51d, discovered by expressed
AID
GENO 5226
/
6r64$$$302
03-13-98 05:33:31
sequence tag (EST) database searches with Rad51 family members (Adams et al., 1995; Lennon et al., 1996). The predicted proteins share similarity with other RAD51-like proteins, suggesting that they have a role in recombination and DNA repair. Here we describe the initial characterization and chromosomal mapping of the mouse and human Rad51d genes. MATERIALS AND METHODS Clone identification. To identify new potential mouse and human homologs of the RAD51-gene family, the S. cerevisiae RAD55 (Lovett, 1994) and human REC2/RAD51B (Rice et al., 1997; Albala et al., 1998) amino acid sequences were used to search the NCBI EST database on the XREF server (Bassett et al., 1995, 1997) and the TIGR Human Gene Index. Two new overlapping mouse cDNAs [I.M.A.G.E. Consortium (LLNL) cDNA clones 746795 and 478233] and two new overlapping human cDNAs were identified [I.M.A.G.E. Consortium (LLNL) cDNA Clones 60163 and 259579]. The 746795 clone was generated from mouse whole fetus, 12.5 dpc, and the 478233 clone was identified from 13.5- to 14.5-dpc whole mouse fetus. The 60163 clone was from human adult T-lymphocytes, and the 259579 clone was from 8- to 9-week-pc human placenta. The cDNA clones were obtained from either American Type Culture Collection (Rockville, MD), or Research Genetics (Huntsville, AL). The 259579, 746795, and 478233 cDNA clones were in the EcoRI and NotI sites of the pT7T3D-Pac vector (Pharmacia) (Lennon et al., 1996). The 60163 clone was in the EcoRI and XhoI sites of the pBluescript SK0 vector (Stratagene). Sequencing, informatics, and nomenclature. Sequencing was performed on a 373 automated sequencer (ABI) with Taq Dye-Deoxy Terminators (ABI), and all clones were completely sequenced on both strands. To confirm the nucleotide sequence of the human RAD51D gene, the oligonucleotide primer pair RAD51D-5* [(5*-CGAGGAGATGATCCAGCTTCTCA-3*) corresponding to /39 bp (see Fig. 1B)] and RAD51D-3* [(5*-CTCCAGTATAGAGACCAGCATCAA-3*) corresponding to /184 bp] was used to amplify DNA extracted from a human neuronal cDNA library constructed from a retinoic acid-induced NT2 cell line (Ackerman et al., 1994) as follows. Following a 2-min denaturing step at 977C, 35 rounds of amplification were performed: 947C for 30 s, 607C for 30 s, and 727C for 1 min. Both strands of the resulting amplimer were sequenced using the same primers. The nucleotide sequence was assembled using the Sequencer 3.0 software (Gene Codes Corp., Ann Arbor, MI). Amino acid and nucleotide comparisons were performed using the GeneWorks 2.4N sequence analysis program (IntelliGenetics; Mountain View, CA), and gaps in alignments were counted as one mismatch. Database searches and sequence comparisons were performed at NCBI (USA) and GenomeNet (Japan). pI and molecular weights were calculated using the pI/MW tool available on the ExPasY molecular biology server (Bjellqvist et al., 1994). This is the fourth RAD51-like gene identified in human based upon sequence similarity to the yeast RAD51 gene. In keeping with the naming system established by the Human Nomenclature Committee, this new gene has been given the name RAD51D in human and Rad51d in mouse. The gene sequences were assigned Accession Nos. AF034955 (Rad51d) and AF034956 (RAD51D). Northern blot analysis. A mouse multiple tissue Northern (MTN) blot (Clontech 7762-1) was hybridized with a mouse Rad51d probe (897-bp EcoRI/SacI fragment from clone 478233) for 1 h at 657C in ExpressHyb (Clontech). Two room-temperature washes in 11 SSC (150 mM NaCl, 15 mM sodium citrate, pH 7), 0.1% SDS were performed for 10 min, followed by one wash at 507C in 0.11 SSC, 0.1% SDS for 20 min. The filter was exposed to a phosphorimaging screen and analyzed on a Fuji BAS 2000 analysis system. The filter was stripped in 0.5% SDS (90–1007C) for 20 min. The human b-actin gene (Clontech) was then used to probe the same filter. Hybridization and wash conditions were the same as described above, except the two final washes were at 657C in 0.11 SSC, 0.1% SDS for 20 min.
gnma
IDENTIFICATION OF Rad51d Zoo blot. A Southern blot filter (Clontech 7753-1), containing 4 mg of EcoRI-digested genomic DNA from human, monkey, rat, mouse, dog, cow, rabbit, chicken, and yeast was hybridized at 657C in Church buffer (Church and Gilbert, 1984) with an 897-bp EcoRI–SacI fragment from the 478233 mouse cDNA clone. Two washes were performed in 11 SSC, 0.1% SDS at room temperature for 10 min and once at 657C 11 SSC, 0.1% SDS for 10 min. Signal was detected on a Fuji BAS 2000 system as described above. Genetic mapping of Rad51d in mouse and human. Mouse Rad51d was mapped on the BSS (C57BL/6J 1 Mus spretus) 1 M. spretus interspecific backcross mapping panel maintained at The Jackson Laboratory as follows. A restriction site polymorphism between the parental strains was identified in a 457-bp DNA fragment generated by PCR amplification at the 3* end of Rad51d. The C57BL/6J allele, but not the M. spretus allele, was cleavable by MspI to yield fragment sizes of 377 and 80 bp. The 5* and 3* primers were 478233.4L (5*GCCCAGAATTACCTGGCAAG-3*) and 478UTR.1241 (5*-GCAGATGGGAAAACAGACCA-3*), respectively. Following a 3-min denaturing step at 947C, 35 rounds of amplification were performed: 947C for 45 s, 557C for 1 min, and 727C for 2 min. The amplification products were then digested with MspI prior to agarose gel electrophoresis. Mapping data and further information on the backcross panel can be found on the World Wide Web at http://www.jax.org/resources/ documents/cmdata/. To map human RAD51D, oligonucleotide primer pair 60163.PCR1 (5*-CTGGCGACTGGATGGATAATC-3*) and 60163.PCR2 (5*-CCCAGTAACTCAGAGACAGAG-3*) was used. Following a 2-min denaturing step at 977C, 35 rounds of amplification were performed: 947C for 30 s, 557C for 30 s, 727C for 1 min. No signal was detected from hamster genomic DNA using these amplification conditions. A 185bp amplimer from the 3* end of RAD51D (1179–1364) was generated from human genomic DNA, and a panel of hamster–human hybrids was screened (Stanford Radiation Hybrid mapping panel G3, Research Genetics). Data were submitted to the Stanford Radiation Hybrid Server ([email protected]), which returned the linkage data reported under Results.
RESULTS
Identification of a New RAD51-Related Gene in Mouse and Human As part of an effort to identify new mammalian genes required for recombination, the amino acid sequences of S. cerevisiae RAD55 (Lovett, 1994) and human RAD51B/REC2 (Rice et al., 1997; Albala et al., 1997) were used to search the NCBI EST database, using the XREF Server (Bassett et al., 1995, 1996, 1997) and the TIGR Human Gene Index. An EST, I.M.A.G.E. Consortium clone ID 478233 (GenBank Accession No. AA049653), was identified, having a match of P value 2.7 1 1003 to RAD55 and 1.1 1 1004 to human RAD51B/REC2. The complete DNA sequence of this clone, which originated from a 13.5- to 14.5-day mouse embryo cDNA library, was determined and used to search the EST division of GenBank. An additional I.M.A.G.E. clone extending 404 bp further 5* was identified from a 12.5-day mouse embryonic library (clone ID 746795; GenBank Accession No. AA260430). Identical matches to this new mouse gene were I.M.A.G.E. clones 692411 (GenBank Accession No. AA237779) and 849350 (GenBank Accession No. AA457827), originating from mouse liver and mammary gland cDNA libraries, respectively. Due to its significant similarity to other RAD51 family members, this new mouse gene has been named Rad51d (see Materials and Methods).
AID
GENO 5226
/
6r64$$$302
03-13-98 05:33:31
105
Sequence analysis suggests that the 746795 mouse cDNA clone is full length (Fig. 1). Translation of a 987bp ORF predicts a 329-amino-acid protein with a molecular mass of 35,260 Da and a pI of 6.76. The sequence surrounding the presumed start codon has optimal context for a translation initiation site, an A at 03 and G at /4, and in-frame stop codons are present upstream of the presumed start codon (Kozak, 1996). No AATAAA polyadenylation signals were present at the 3* end, but a poly(A) tail was present on both Rad51d EST clones. Further GenBank queries with the mouse Rad51d nucleotide sequence revealed two highly significant hits with human ESTs, I.M.A.G.E. Clone 60163 (GenBank Accession No. AA352205) from human T-lymphocytes and Clone 259579 (GenBank Accession No. N57184) from human placenta. Both appear to be full length in the sense that they contain the initiation codon. Overall, the two clones showed extensive sequence similarity to mouse Rad51d with similar predicted start and stop sites (Fig. 1). However, each has unusual structural differences. The shorter of the two cDNA clones, 259579, appears to represent an alternatively spliced product; it excludes sequences corresponding to two exons in the mouse gene (D. Pittman, unpublished observation), one encoding the first ATP-binding motif (see Fig. 1 legend), and encodes a 216-amino-acid protein. The 60163 EST clone also contains an apparent alternative splice that skips a single exon present in the mouse gene (D. Pittman, unpublished observation; see Fig. 1). The absence of this exon results in the introduction of a premature stop codon due to a frameshift, thereby predicting a truncated protein containing only 49 amino acids. To exclude the possibility that the frameshift in 60163 was not a cloning artifact, a PCR product encompassing these nucleotides was generated from a human neuronal cDNA library (see Materials and Methods). The nucleotide sequence was identical to the sequence of the 60163 EST clone. If the frameshift is ignored (Fig. 1B), the 60163 insert would encode a 289-amino-acid protein sharing 71% overall amino acid identity and 73% overall nucleotide identity with the mouse gene, suggesting that the human and mouse genes are orthologs (see also syntenic mapping data below). Accordingly, the human gene has been assigned the name RAD51D (see Materials and Methods). The theoretical molecular mass of the protein encoded by the larger human cDNA clone, 60163, is 30,960 Da with a pI of 5.87. Similarity of Rad51d/RAD51D to the RAD51 Family of Genes Searches of the databases using the S. cerevisiae Rad51 protein produced an extremely large number of significant hits with other RAD51 homologs, obscuring more weakly related genes. Even though the original Rad51d clone, 478233, was identified in a search against RAD55, Rad51d shows the highest level of
gnma
106
PITTMAN, WEINBERG, AND SCHIMENTI
amino acid similarity to RAD51 genes present in several organisms (e.g., yeast, chick, mouse, and human). This relatedness is illustrated in Fig. 1C by alignment with the mouse Rad51a member of the RAD51 gene family. Like the other members of the RAD51 gene family, mouse and human Rad51d/RAD51D genes have two conserved ATP-binding domains (Fig. 1). The first (GGPGSGKTQ) is located between amino acids 107 and 115 in the mouse sequence and between amino acids 67 and 75 in the human sequence. Another highly conserved region is a second ATP-binding domain (VVVIVDS) present at amino acids 202–207 in the Rad51d sequence and at amino acids 163–168 in the human RAD51D protein. Finally, the residues NH at positions 249–250 in Rad51d (and also in RAD51D), which are conserved as either NH or NQ in other RecArelated genes and are thought to be involved in ATPinduced conformational change, are present in these genes (Story et al., 1993). Conservation of Rad51d To determine the degree of conservation of Rad51d in other mammalian species, a Southern blot of genomic DNA from seven mammalian species was hybridized using Rad51d (an 897-bp EcoRI/SacI fragment from the 478233 clone) as the probe (Fig. 2). Strong signals were detected in DNA from all mammalian species tested and from yeast (S. cerevisiae). The signal detected from yeast genomic DNA is most likely RAD51 due to its similarity to Rad51d (19% identity at the amino acid level). The size of the fragment is also consistent with the size predicted from the genomic sequence around RAD51, approximately 6.1 kb. A weak signal was also detected in chicken. These results indicated conservation of the Rad51d gene across mammalian species. Rad51d Tissue Distribution To determine the size and tissue distribution of the Rad51d transcripts in mouse, a Northern blot, containing poly(A)/ RNA from several mouse tissues, was hybridized with a 32P-labeled Rad51d probe (Fig. 3). Multiple transcripts, 7.7, 2.5, and 1.7 kb, were detected in every tissue examined. The smaller transcript is consistent with the size of the mouse cDNA clones. The 7.7-kb transcript appears to be the strongest signal in brain tissue, and the two smaller transcripts appear to be strongest in all other tissues. A 3.6-kb testis-specific transcript was also detected. Rad51d Maps to Mouse Chromosome 11 and RAD51D Maps to Human Chromosome 17 Using a restriction site difference between parental strains of The Jackson Laboratory’s interspecific mouse
backcross (see Materials and Methods), Rad51d was mapped to position 48.5 (according to the Mouse Genome Database) on Chromosome 11 (lod score 25.9), in an interval that also contains the microsatellite marker D11Mit36 (Fig. 4). A scan of the region for mapped traits that might be consistent with a mutation in Rad51d revealed the pulmonary adenoma resistance gene (Par1). Par1 is a quantitative trait locus (QTL) in M. spretus that confers partial resistance to lung tumor susceptibility present in the strain A/J (Manenti et al., 1996). The chromosomal location of human RAD51D was identified using the Stanford Radiation Hybrid mapping panel G3. The human RAD51D gene was mapped within 7.7 cR from STS WI-2760 (aka, D17S822E, EST00213) with a lod score of 13.6 (Fig. 5). The WI2760 marker is located at 58–65 cM on chromosome 17, placing RAD51D at 17q11 within a 7-cM region flanked by markers D17S933 and D17S791. These regions of mouse Chromosome 11 and human chromosome 17 share synteny, suggesting that these two genes are direct orthologs. DISCUSSION
Recombinational Repair and the RecA-Related Genes Systematic identification of eukaryotic genes involved in recombinational repair has been conducted predominantly in the yeast S. cerevesiae. Of the numerous experimental advantages afforded by this organism, the ability to screen haploid cells for recessive mutations was particularly important in the isolation of repair mutants. Because ionizing radiation causes DNA double-stranded breaks, and double-stranded breaks are repaired by a recombinational mechanism, several mutations isolated in screens for sensitivity to X rays ultimately proved to affect genes involved in recombination. RAD51 was one of several yeast genes identified in screens for X-ray-sensitive mutants (Game and Mortimer, 1974; Game, 1993). RAD51 is related to the E. coli RecA protein, which is involved in homologous recombination and DNA repair (see Edelmann and Kucherlapati, 1996). Indeed, rad51 mutants are defective in mitotic recombination, indicative of Rad51’s role in a recombination repair system. RAD55 and RAD57 are two additional yeast genes required for efficient repair of ionizing radiation-induced damage that share homology to RecA, and rad55 and rad57 mutants are also recombination-defective (Game and Mortimer, 1974; Lovett and Mortimer, 1987). Several lines of evidence point to the existence of
FIG. 1. Nucleotide and predicted amino acid sequences of (A) mouse Rad51d and (B) human RAD51D. The two conserved nucleotide binding domains are underlined. The position of an apparent frameshift, due to the possible absence of a single exon, in human I.M.A.G.E. clone 60163 is indicated by NN at nucleotide position 146 (see text). Note that the I.M.A.G.E. clone 259579 does not contain basepairs 145 to 363. (C) Amino acid alignment of the mouse and human Rad51d proteins along with the amino acid sequence of mouse Rad51a. (Dots represent identical amino acids; dashes indicate gaps in the amino acid alignment.)
AID
GENO 5226
/
6r64$$$303
03-13-98 05:33:31
gnma
IDENTIFICATION OF Rad51d
AID
GENO 5226
/
6r64$$5226
03-13-98 05:33:31
gnma
107
108
PITTMAN, WEINBERG, AND SCHIMENTI
sis-specific). In addition to having orthologs of RAD51 and DMC1, at least three other homologs have been discovered in mice (and humans): RAD51B/Rec2, RAD51C, and Rad51d/RAD51D. It is not clear what the orthologs of these last three genes are, whether they are direct relatives of RAD51 itself or of DMC1, RAD55, and RAD57. The situation may be clarified as additional RecA homologs are identified or as the actual biochemical functions of the genes are elucidated. A truncated form of RAD51D in humans was identified from cDNAs generated from two different tissues. A second splice variant that lacked the nucleotide-binding motif critical for the function of RAD51 was identified. These alternatively spliced variants raise some interesting questions as to whether the peptide made from the latter transcript is functional and whether posttranscriptional regulation is occurring at the level of RNA processing. A mutation in the ATP-binding residue in the yeast RAD57 gene did not lead to loss of activity (Johnson and Symington, 1995). Perhaps this motif is not critical for the function of RAD51D or lack of this motif is complemented by another RAD51 family member present in a recombinational complex of proteins. Isolation of additional RAD51D cDNAs and the FIG. 2. Conservation of Rad51d. An 897-bp EcoRI/SacI fragment from the mouse 478233 cDNA clone was used to probe a Southern blot containing approximately 4 mg of EcoRI-digested genomic DNA from several mammalian species, chicken, and yeast. Size markers are indicated on the left.
a recombinational repair system in mammals. First, human and mouse homologs of fungal recombinational repair genes have recently been identified. These include RAD51A (Shinohara et al., 1993; Morita et al., 1993), RAD52 (Bendixen, 1994; Muris et al., 1994; Shen et al., 1995), Rec2/RAD51B (Rice et al., 1997; Albala et al., 1997), which is a RAD51 homolog, RAD51C (a Rad51-related gene isolated in the laboratory of D. Schild, pers. comm., Berkeley, CA, October 1997), and RAD51D. Second, as with yeast RAD51, strand exchange activity has been demonstrated for the human RAD51 protein, indicating conserved function (Baumann et al., 1996; Ogawa et al., 1993; Sung, 1994). Third, site-specific double-strand breaks induced in mammalian cells by a rare-cutting endonuclease can be repaired by homologous recombination in addition to end-joining (Rouet et al., 1994; Sargent et al., 1997). However, the majority of stimulated repair events are nonhomologous, indicating that repair of double-strand breaks by recombination is not the major pathway in mitotic cells. Fourth, mammalian homologs of the yeast RAD genes have been shown to interact (see Golub et al., 1997). It appears that mammals may have an expanded RecA family, relative to yeast. Because the sequence of the S. cerevesiae genome has been determined, we know that yeast contains four RecA-related genes: RAD51, RAD55, RAD57 and DMC1 (the latter is meio-
AID
GENO 5226
/
6r64$$$304
03-13-98 05:33:31
FIG. 3. Expression analysis of mouse Rad51d. (Top) A mouse tissue blot was probed with an 897-bp EcoRI/SacI fragment from the I.M.A.G.E. clone 478233. The Rad51d transcript sizes are indicated on the right. (Bottom) A b-actin cDNA probe was used to verify the presence of RNA in each lane.
gnma
IDENTIFICATION OF Rad51d
109
Whether Rad51d is involved in either of these genetic disorders requires further study. Recombinational Repair Genes and Meiosis Genetic control of recombination in mammals has been exceedingly difficult to study, because tools do not exist to screen efficiently for mutations that may result in meiotic failure (hence, sterility). However, investigations in yeast have shown that several proteins crucial for meiotic recombination are also involved in mitotic recombination and vice versa. Mutations in RAD51, RAD55, and RAD57 affect meiotic as well as mitotic recombination (Game, 1993). It is reasonable to assume that the same will hold true for mammalian recombinational repair genes. Therefore, in the pursuit of meiotic recombination genes, a potentially fruitful, more technically feasible approach may be to identify genes involved in mitotic recombinational repair. One obvious avenue is to screen the EST database for homologs of known yeast recombination genes, as we have accomplished here. There is also a practical advantage to this strategy: large-scale EST sequencing efforts have yet to conduct extensive sequencing of germ-cell cDNA libraries. Furthermore, although clones from entire re-
FIG. 4. Localization of Rad51d on mouse Chromosome 11. The gene order and relative marker positions are shown. Map distance in centimorgans is shown on the left.
genomic sequence will help resolve these questions. Furthermore, despite the overall similarity to Rad51a, it is not certain whether Rad51d encodes similar biochemical function. For example, yeast RAD55 and RAD57 are not directly involved in strand exchange, but serve as accessory proteins to stabilize or enhance the activity of RAD51 (Sung, 1997). Should Rad51d be involved in DNA repair, then mutations in this gene might cause phenotypes such as tumor susceptibility or genome instability. Inspection of the mouse Chromosome 11 genetic map near Rad51d revealed close linkage to a QTL for pulmonary adenoma resistance (Par1) (Manenti et al., 1996). The A/J strain of mice has a susceptible allele at this locus, whereas the M. spretus allele confers resistance in heterozygotes. Furthermore, the human homolog (RAD51D) maps to 17q11.2, near the FWT1 gene that confers Wilms tumor susceptibility (Rahman et al., 1996). This gene has been mapped in an interval (17q12–q21) that is apparently just distal to RAD51D.
AID
GENO 5226
/
6r64$$$304
03-13-98 05:33:31
FIG. 5. Chromosomal localization of human RAD51D. Idiogram of human chromosome 17 showing linkage of RAD51D to D17S822E as determined by radiation hybrid mapping using the G3 Stanford Radiation Hybrid Panel (1 cR equals approximately 30 kb). Distances, in cR, between outside markers, and lod scores are presented. Map distances were taken from the Stanford RH Mapping Index at http://www-shgc.stanford.edu/Mapping/rh/.
gnma
110
PITTMAN, WEINBERG, AND SCHIMENTI
productive organs are being sequenced, meiotic prophase staged cells represent only a subset of all cell types. Hence, genes expressed at relatively low levels at specific meiotic stages may escape sequence determination. Nevertheless, RAD51-related genes in mammals have been identified using degenerate PCR approaches, as in the case of REC2/RAD51B, RAD51, and DMC1 (Rice et al., 1997; Morita et al., 1993; Shinohara et al., 1993; Sato et al., 1995a). Current evidence suggests roles for mammalian RAD51A in meiotic recombination. Mouse Rad51a is expressed at high levels in ovary and testis, and the protein is associated with the axial/lateral element in synaptonemal complexes in mouse spermatocytes and oocytes (Shinohara et al., 1993; Ashley et al., 1995; Haaf et al., 1995; Plug et al., 1996). Unfortunately, targeted mutagenesis of Rad51a resulted in early embryonic lethality, precluding a study of its possible role in meiosis (Tsuzuki et al., 1996). It is possible that similar problems may be encountered with targeted mutants of the other RAD51-related recombinational repair genes. In light of this, it appears that conditional mutations may be required to understand the in vivo roles of these genes in meiotic vs mitotic cells. Additionally, the possible effects of mutations in these genes on recombination may require assays, possibly employing transgenes or PCR, capable of measuring gene conversion and crossing over in situations where infertility may result. ACKNOWLEDGMENTS The authors thank Mary Barter and Lucy Rowe for assistance in mapping Rad51d using the BSS backcross, Doug McMinimy and Amy Lambert for automated sequencing, Patsy Nishina for providing DNA from the recombination hybrid mapping panel, Joyce Worcester for preparing figures, David Schild and Larry Thompson for providing unpublished information, Sue Ackerman for providing the human cDNA library, and Patsy Nishina and Tim O’Brien for critically reading the manuscript. This work was supported by NIH Grant GM45415 to J.S. and by a Cancer Center core grant (CA34196) to The Jackson Laboratory. D.P. was supported by NIH Training Grant HD07065 from the National Institute of Child Health and Human Development, by a fellowship from the Lalor Foundation, and by NRSA Award GM19052.
REFERENCES Ackerman, S. L., Knowles, B. B., and Andrews, P. W. (1994). Gene regulation during neuronal and non-neuronal differentiation of NTERA2 human teratocarcinoma-derived stem cells. Brain Res. Mol. Brain Res. 25: 157–162. Adams, M. D., Kerlavage, A. R., Fleischmann, R. D., Fuldner, R. A., Bult, C. J., Lee, N. H., Kirkness, E. F., Weinstock, K. G., Gocayne, J. D., White, O., et al. (1995). Initial assessment of human gene diversity and expression patterns based upon 83 million nucleotides of cDNA sequence. Nature 377: 3–174. Albala, J. S., Thelen, M. P., Prange, C., Fan, W., Christensen, M., Thompson, L. H., and Lennon, G. G. (1998). Identification of a novel human RAD51 homolog, RAD51B. Genomics, in press. Ashley, T., Plug, A. W., Xu, J., Solari, A. J., Reddy, G., Golub, E. I., and Ward, D. C. (1995). Dynamic changes in Rad51 distribution
AID
GENO 5226
/
6r64$$$304
03-13-98 05:33:31
on chromatin during meiosis in male and female vertebrates. Chromosoma 104: 19–28. Bassett, D. E., Jr., Boguski, M. S., and Hieter, P. (1996). Yeast genes and human disease. Nature 379: 589–590. [Letter] Bassett, D. E., Jr., Boguski, M. S., Spencer, F., Reeves, R., Goebl, M., and Hieter, P. (1995). Comparative genomics, genome crossreferencing and XREFdb. Trends Genet. 11: 372–373. Bassett, D. E., Jr., Boguski, M. S., Spencer, F., Reeves, R., Kim, S., Weaver, T., and Hieter, P. (1997). Genome cross-referencing and XREFdb: Implications for the identification and analysis of genes mutated in human disease. Nat. Genet. 15: 339–344. Baumann, P., Benson, F. E., and West, S. C. (1996). Human Rad51 protein promotes ATP-dependent homologous pairing and strand transfer reactions in vitro. Cell 87: 757–766. Bendixen, C., Sunjevaric, I., Bauchwitz, R., and Rothstein, R. (1994). Identification of a mouse homologue of the Saccharomyces cerevisiae recombination and repair gene, RAD52. Genomics 23: 300– 303. Bishop, D. K. (1994). RecA homologs Dmc1 and Rad51 interact to form multiple nuclear complexes prior to meiotic chromosome synapsis. Cell 79: 1081–1092. Bishop, D. K., Park, D., Xu, L., and Kleckner, N. (1992). DMC1: A meiosis-specific yeast homolog of E. coli recA required for recombination, synaptonemal complex formation, and cell cycle progression. Cell 69: 439–456. Bjellqvist, B., Basse, B., Olsen, E., and Celis, J. E. (1994). Reference points for comparisons of two-dimensional maps of proteins from different human cell types defined in a pH scale where isoelectric points correlate with polypeptide compositions. Electrophoresis 15: 529–539. Bronner, C. E., Baker, S. M., Morrison, P. T., Warren, G., Smith, L. G., Lescoe, M. K., Kane, M., Earabino, C., Lipford, J., Lindblom, A., et al. (1994). Mutation in the DNA mismatch repair gene homologue hMLH1 is associated with hereditary non-polyposis colon cancer. Nature 368: 258–261. Church, G., and Gilbert, W. (1984). Genomic sequencing. Proc. Natl. Acad. Sci. USA 81: 1991–1995. Clever, B. I. H., Schmuckli-Maurer, J., King, J., Sigrist, M., and Heyer, W. D. (1997). Recombinational repair in yeast: Functional interactions between Rad51 and Rad54 proteins. EMBO J. 16: 2535–2544. Donovan, J. W., Milne, G. T., and Weaver, D. T. (1994). Homotypic and heterotypic protein associations control Rad51 function in double-strand break repair. Genes Dev. 8: 2552–2562. Edelmann, W., and Kucherlapati, R. (1996). Role of recombination enzymes in mammalian cell survival. Proc. Natl. Acad. Sci. USA 93: 6225–6227. Fishel, R., Lescoe, M. K., Rao, M. R., Copeland, N. G., Jenkins, N. A., Garber, J., Kane, M., and Kolodner, R. (1993). The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer. Cell 75: 1027–1038. Game, J. C. (1993). DNA double-strand breaks and the RAD50RAD57 genes in Saccharomyces. Cancer Biol. 4: 73–83. Game, J. C., and Mortimer, R. K. (1974). A genetic study of X-ray sensitive mutants in yeast. Mutat. Res. 24: 281–292. Golub, E. I., Kovalenko, O. V., Gupta, R. C., Ward, D. C., and Radding, C. M. (1997). Interaction of human recombination proteins Rad51 and Rad54. Nucleic Acids Res. 25: 4106–4110. Gupta, R. C., Bazemore, L. R., Golub, E. I., and Radding, C. M. (1997). Activities of human recombination protein Rad51. Proc. Natl. Acad. Sci. USA 94: 463–468. Haaf, T., Golub, E. I., Reddy, G., Radding, C. M., and Ward, D. C. (1995). Nuclear foci of mammalian Rad51 recombination protein in somatic cells after DNA damage and its localization in synaptonemal complexes. Proc. Natl. Acad. Sci. USA 92: 2298–2302. Habu, T., Taki, T., West, A., Nishimune, Y., and Morita, T. (1996). The mouse and human homologs of DMC1, the yeast meiosis-spe-
gnma
IDENTIFICATION OF Rad51d cific homologous recombination gene, have a common unique form of exon-skipped transcript in meiosis. Nucleic Acids Res. 24: 470– 477. Hays, S. L., Firmenich, A. A., and Berg, P. (1995). Complex formation in yeast double-strand break repair: Participation of Rad51, Rad52, Rad55, and Rad57 proteins. Proc. Natl. Acad. Sci. USA 92: 6925–6929. Johnson, R. D., and Symington, L. S. (1995). Functional differences and interactions among the putative RecA homologs Rad51, Rad55, and Rad57. Mol. Cell. Biol. 15: 4843–4850. Kozak, M. (1996). Interpreting cDNA sequences: Some insights from studies on translation. Mamm. Genome 7: 563–574. Lamb, N. E., Freeman, S. B., Savage Austin, A., Pettay, D., Taft, L., Hersey, J., Gu, Y., Shen, J., Saker, D., May, K. M., Avramopoulos, D., Petersen, M. B., Hallberg, A., Mikkelsen, M., Hassold, T. J., and Sherman, S. L. (1996). Susceptible chiasmate configurations of chromosome 21 predispose to non-disjunction in both maternal meiosis I and meiosis II. Nat. Genet. 14: 400–405. Lennon, G., Auffray, C., Polymeropoulos, M., and Soares, M. B. (1996). The I.M.A.G.E. consortium: An integrated molecular analysis of genomes and their expression. Genomics 33: 151–152. Lovett, S. T. (1994). Sequence of the RAD55 gene of Saccharomyces cerevisiae: Similarity of RAD55 to prokaryotic RecA and other RecA-like proteins. Gene 142: 103–106. Lovett, S. T., and Mortimer, R. K. (1987). Characterization of null mutants of the RAD55 gene of Saccharomyces cerevisiae: Effects of temperature, osmotic strength and mating type. Genetics 116: 547–553. Manenti, G., Gariboldi, M., Elango, R., Fiorino, A., De Gregorio, L., Falvella, F. S., Hunter, K., Housman, D., Pierotti, M. A., and Dragani, T. A. (1996). Genetic mapping of a pulmonary adenoma resistance (Par1) in mouse. Nat. Genet. 12: 455–457. Milne, G. T., and Weaver, D. T. (1993). Dominant negative alleles of RAD52 reveal a DNA repair/recombination complex including Rad51 and Rad52. Genes Dev. 7: 1755–1765. Morita, T., Yoshimura, Y., Yamamoto, A., Murata, K., Mori, M., Yamamoto, H., and Matsushiro, A. (1993). A mouse homolog of the Escherichia coli recA and Saccharomyces cerevisiae RAD51 genes. Proc. Natl. Acad. Sci. USA 90: 6577–6580. Muris, D. F., Bezzubova, O., Buerstedde, J. M., Vreeken, K., Balajee, A. S., Osgood, C. J., Troelstra, C., Hoeijmakers, J. H., Ostermann, K., Schmidt, H., et al. (1994). Cloning of human and mouse genes homologous to RAD52, a yeast gene involved in DNA repair and recombination. Mutat. Res. 315: 295–305. Nicolaides, N. C., Papadopoulos, N., Liu, B., Wei, Y. F., Carter, K. C., Ruben, S. M., Rosen, C. A., Haseltine, W. A., Fleischmann, R. D., Fraser, C. M., et al. (1994). Mutations of two PMS homologues in hereditary nonpolyposis colon cancer. Nature 371: 75–80. Ogawa, T., Yu, X., Shinohara, A., and Egelman, E. H. (1993). Similarity of the yeast RAD51 filament to the bacterial RecA filament. Science 259: 1896–1899. Plug, A. W., Xu, J., Reddy, G., Golub, E. I., and Ashley, T. (1996).
AID
GENO 5226
/
6r64$$$305
03-13-98 05:33:31
111
Presynaptic association of Rad51 protein with selected sites in meiotic chromatin. Proc. Natl. Acad. Sci. USA 93: 5920–5924. Radding, C. M. (1991). Helical interactions in homologous pairing and strand exchange driven by RecA protein. J. Biol. Chem. 266: 5355–5358. Rahman, N., Arbour, L., Tonin, P., Renshaw, J., Pelletier, J., Baruchel, S., Pritchard Jones, K., Stratton, M. R., and Narod, S. A. (1996). Evidence for a familial Wilms’ tumour gene (FWT1) on chromosome 17q12–q21. Nat. Genet. 13: 461–463. Rice, M. C., Smith, S. T., Bullrich, F., Havre, P., and Kmiec, E. B. (1997). Isolation of human and mouse genes based on homology to REC2, a recombinational repair gene from the fungus Ustilago maydis. Proc. Natl. Acad. Sci. USA 94: 7417–7422. Rouet, P., Smih, F., and Jasin, M. (1994). Expression of a site-specific endonuclease stimulates homologous recombination in mammalian cells. Proc. Natl. Acad. Sci. USA 91: 6064–6068. Sargent, R. G., Brenneman, M. A., and Wilson, J. H. (1997). Repair of site-specific double-strand breaks in a mammalian chromosome by homologous and illegitimate recombination. Mol. Cell. Biol. 17: 267–277. Sato, S., Kobayashi, T., Hotta, Y., and Tabata, S. (1995a). Characterization of a mouse recA-like gene specifically expressed in testis. DNA Res. 2: 147–150. Sato, S., Seki, N., Hotta, Y., and Tabata, S. (1995b). Expression profiles of a human gene identified as a structural homologue of meiosis-specific recA-like genes. DNA Res. 2: 183–186. Shen, Z., Denison, K., Lobb, R., Gatewood, J. M., and Chen, D. J. (1995). The human and mouse homologs of the yeast RAD52 gene: cDNA cloning, sequence analysis, assignment to human chromosome 12p12.2–p13, and mRNA expression in mouse tissues. Genomics 25: 199–206. Shinohara, A., Ogawa, H., Matsuda, Y., Ushio, N., Ikeo, K., and Ogawa, T. (1993). Cloning of human, mouse and fission yeast recombination genes homologous to RAD51 and recA. Nat. Genet. 4: 239–243. Shinohara, A., Ogawa, H., and Ogawa, T. (1992). Rad51 protein involved in repair and recombination in S. cerevisiae is a RecA-like protein. Cell 69: 457–470. Story, R. M., Bishop, D. K., Kleckner, N., and Steitz, T. A. (1993). Structural relationship of bacterial RecA proteins to recombination proteins from bacteriophage T4 and yeast. Science 259: 1892– 1896. Sung, P. (1994). Catalysis of ATP-dependent pairing and strand exchange by yeast RAD51 protein. Science 265: 1241–1243. Sung, P. (1997). Yeast Rad55 and Rad57 proteins form a heterodimer that functions with replication protein A to promote DNA strand exchange by Rad51 recombinase. Genes Dev. 11: 1111–1121. Tsuzuki, T., Fujii, Y., Sakumi, K., Tominaga, Y., Nakao, K., Sekiguchi, M., Matsushiro, A., Yoshimura, Y., and Morita, T. (1996). Targeted disruption of the Rad51 gene leads to lethality in embryonic mice. Proc. Natl. Acad. Sci. USA 93: 6236–6240. West, S. C. (1992). Enzymes and molecular mechanisms of genetic recombination. Annu. Rev. Biochem. 61: 603–640.
gnma