The N-Terminal DNA-Binding Domain of Rad52 Promotes RAD51-Independent Recombination in Saccharomyces cerevisiae

Corresponding author: Akira Shinohara, Graduate School of Science, Osaka University, 1-1 Machikaneyama, Toyonaka, Osaka 560-0043, Japan., ashino{at}bio.sci.osaka-u.ac.jp (E-mail)

Communicating editor: L. SYMINGTON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In Saccharomyces cerevisiae, the Rad52 protein plays a role in both RAD51-dependent and RAD51-independent recombination pathways. We characterized a rad52 mutant, rad52-329, which lacks the C-terminal Rad51-interacting domain, and studied its role in RAD51-independent recombination. The rad52-329 mutant is completely defective in mating-type switching, but partially proficient in recombination between inverted repeats. We also analyzed the effect of the rad52-329 mutant on telomere recombination. Yeast cells lacking telomerase maintain telomere length by recombination. The rad52-329 mutant is deficient in RAD51-dependent telomere recombination, but is proficient in RAD51-independent telomere recombination. In addition, we examined the roles of other recombination genes in the telomere recombination. The RAD51-independent recombination in the rad52-329 mutant is promoted by a paralogue of Rad52, Rad59. All components of the Rad50-Mre11-Xrs2 complex are also important, but not essential, for RAD51-independent telomere recombination. Interestingly, RAD51 inhibits the RAD51-independent, RAD52-dependent telomere recombination. These findings indicate that Rad52 itself, and more precisely its N-terminal DNA-binding domain, promote an essential reaction in recombination in the absence of RAD51.

HOMOLOGOUS recombination is important in genome stability. Dysfunction results in genome instability, which is often associated with the onset of cancer. Recombination is required for the repair of double-strand breaks (DSBs) as well as for the segregation of homologous chromosomes during meiosis. Recently it has been shown that recombination is also involved in maintaining telomere length.

In Saccharomyces cerevisiae, genes in the RAD52 epistasis group (RAD50, -51, -52, -54, -55, -57, -59, TID1/RDH54, MRE11, and XRS2) are involved in recombination (PAQUES and HABER 1999 ; SYMINGTON 2002 ). So far, two pathways have been identified: RAD51 dependent and RAD51 independent. RAD51 encodes a bacterial RecA homolog, which works during the search for homology and the exchange of strands (SHINOHARA et al. 1992 ; SUNG 1994 ). The RAD51-dependent pathway is a major conservative repair pathway for DSBs (PAQUES and HABER 1999 ), but in some circumstances, the nonconservative RAD51-independent pathway operates. The RAD51-dependent pathway requires most of the members of the RAD52 epistasis group, but the RAD51-independent pathway requires only a few members.

In the absence of RAD51, a single DSB on one chromosome in diploid cells is repaired by RAD51-independent recombination, termed break-induced replication (BIR; MALKOVA et al. 1996 ; KRAUS et al. 2001 ). A centromere-proximal DSB can be repaired by invasion of the broken end into a homologous region on another chromosome (Fig 1). A replication fork can be established, and the entire chromosome arm can be copied, resulting in the duplication of a large portion of the chromosome arm. It depends totally on RAD52 (MALKOVA et al. 1996 ) and also involves the Mre11-Rad50-Xrs2 (MRX) complex, Rad59 (a paralogue of Rad52), and Tid1/Rdh54 (SIGNON et al. 2001 ). In addition, BIR initiated by a DSB at the MAT locus requires a cis-acting sequence (MALKOVA et al. 2001 ). Although BIR was originally described for the rad51 mutant, it also occurs in wild-type (WT) cells (MORROW et al. 1997 ; KRAUS et al. 2001 ). RAD59 was originally identified as a gene necessary for recombination between inverted repeats in cells lacking RAD51 (BAI and SYMINGTON 1996 ), which is thought to occur through a BIR event coupled with single-strand annealing (KANG and SYMINGTON 2000 ).

View larger version (24K): [in this window] [in a new window] [Download PPT slide]   Figure 1. A model for BIR in the absence of RAD51. After the resection of a DSB end, ssDNA is used for homology search and strand exchange, resulting in the formation of a D-loop. The 3' end of the invaded strand becomes a substrate for leading-strand synthesis. How a subsequent lagging strand is synthesized is unknown (see KRAUS et al. 2001 for more detail). A newly synthesized strand is shown as a gray line.

Recombination is necessary for telomere elongation in cells lacking telomerase (KASS-EISLER and GREIDER 2000 ; MCEACHERN et al. 2000 ) and for the contraction of telomeres in wild-type cells (BUCHOLC et al. 2001 ). In S. cerevisiae, telomeres consist of 350 bp of TG_1–3 repeats and subtelomeric Y' sequences (long and short Y') next to the repeats. Telomerase, a reverse transcriptase composed of a catalytic protein, Est2, and other proteins, as well as an RNA component encoded by the TLC1 gene, performs de novo synthesis of the repeats (SINGER and GOTTSCHLING 1994 ). If any components of the telomerase are missing, telomeres are gradually shortened and most cells die. However, survivors arise and can elongate their telomeres by recombination (LUNDBLAD and BLACKBURN 1993 ; SINGER and GOTTSCHLING 1994 ). Two types of survivors are generated (LUNDBLAD and BLACKBURN 1993 ; LE et al. 1999 ; TENG and ZAKIAN 1999 ). Type I survivors often amplify the subtelomeric Y' sequence, whereas type II survivors are generated by amplification of TG_1–3 repeats. Type I survivors use RAD51-dependent recombination, while type II survivors use RAD51-independent recombination (LE et al. 1999 ; TENG and ZAKIAN 1999 ). In this article, these are termed type I and type II telomere recombination, respectively. The mechanism of telomere recombination is proposed to be similar to that of BIR (KASS-EISLER and GREIDER 2000 ; KRAUS et al. 2001 ).

The RAD52 gene is required for most recombination in yeast (SYMINGTON 2002 ). Rad52 consists of at least two domains: the N-terminal DNA-binding domain and the C-terminal Rad51-interacting domain (SHINOHARA et al. 1992 ; MILNE and WEAVER 1993 ; SHINOHARA and OGAWA 1998 ). The N-terminal domain is conserved among species while the C-terminal domain is not (MURIS et al. 1994 ). The Rad59 protein shares homology with the N terminus of Rad52 and lacks the C-terminal domain (BAI and SYMINGTON 1996 ). In vivo, Rad52 and Rad59 form a complex (DAVIS and SYMINGTON 2001 ). However, genetic analysis shows that Rad59 plays a less significant role than Rad52 in recombination. Purified yeast and human Rad52 proteins bind to both single-stranded DNA (ssDNA) and double-stranded DNA (dsDNA; MORTENSEN et al. 1996 ; SHINOHARA et al. 1998 ; PARSONS et al. 2000 ) and form ring-like structures by themselves and on the DNAs (SHINOHARA et al. 1998 ; VAN DYCK et al. 1998 ; STASIAK et al. 2000 ; KAGAWA et al. 2002 ; SINGLETON et al. 2002 ). The N-terminal domain of Rad52 is responsible for the DNA binding and ring formation. Rad52 stimulates Rad51-mediated recombination (SUNG 1997 ; NEW et al. 1998 ; SHINOHARA and OGAWA 1998 ) and annealing of complementary ssDNAs in vitro (MORTENSEN et al. 1996 ; SHINOHARA et al. 1998 ; SUGIYAMA et al. 1998 ).

Here, we show that the N-terminal DNA-binding domain of Rad52 facilitates RAD51-independent recombination at telomeres. Genetic analyses using various mutants support the view that the telomere recombination is mechanistically related to BIR. The RAD51-independent recombination pathway is stimulated by the elimination of RAD51.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains and plasmids:
All yeast strains and their genotypes are shown in Table 1. Strains used for telomere analysis (except JHUY564; kindly provided by Carol Greider) were derivatives of S288C. Recombination between inverted repeats was measured in the background of W303. The rad50, rad51, rad52, rad59, mre11, and xrs2 deletion strains were constructed by one-step gene replacement using a fragment containing rad50::hisG-URA3-hisG (ALANI et al. 1989 ), rad51::hisG-URA3-hisG (SHINOHARA et al. 1992 ), rad52::hisG-URA3-hisG (a gift from T. Ogawa), rad59::hisG-URA3-hisG, mre11::hisG-URA3-hisG (JOHZUKA and OGAWA 1995 ), and xrs2::KanMX6, respectively. rad59::hisG-URA3-hisG was constructed by inserting a BamHI-BglII fragment containing hisG-URA3-hisG from pNKY51 (ALANI et al. 1987 ) into the BglII-Eco47III site of the RAD59 gene. xrs2::KanMX6 was constructed as follows: A fragment containing the XRS2 promoter was amplified by PCR using primers (X2-1, 5'-CCGCTCGAGAGAGGACACCAAAG, and X2-2, 5'-GTACTACCCACATATGTTTATAGTTATC) and cloned into the XhoI-EcoRV site of pBluescript II SK+ (pMS235). A PstI-BamHI fragment of YCp50-XRS2 (a gift from Jim Haber) was inserted into the PstI-BamHI site of pMS235, resulting in the plasmid pMS236. The plasmid pMS237 was constructed by cloning a NdeI-SpeI fragment containing KanMX6 from pMJ476 (WACH et al. 1994 ) into the NdeI-SpeI site of pMS236. The rad52-329 allele was constructed by replacing a BamHI-SalI fragment of the RAD52 gene by a PCR fragment amplified using the following oligonucleotides: oligo(A) (5'-CGggatccCTGAAACGCTTCCTGGCCG), which contains a termination codon (underlined) next to the BamHI site followed by the sequence downstream from the authentic stop codon of RAD52, and oligo(B) (5'-GGgtcgacGTCCAAGAAATACATTGG), which contains the SalI site downstream of RAD52. A fragment containing the rad52-329 allele is inserted into the SmaI-EcoRI site of YIplac195 (GIETZ and SUGINO 1988 ), resulting in the plasmid pMT001. Parental strains were transformed with pMT001 digested with BstEII and selected on a synthetic dextrose (SD) medium plate lacking uracil. Ura⁺ transformants were picked and then selected on a plate containing 5-fluoroorotic acid for Ura^-. PCR and Southern blotting confirmed genotypes of the mutants.

The rad52-329 and rad51 deletion alleles were introduced into YDB057 or yAR071 as described above. The rad52 deletion was introduced into yAR71 by adaptamer-mediated PCR using the KlURA3 gene (REID et al. 2002 ).

Liquid assay for telomere recombination:
Diploid cells were transformed with a tlc1::LEU2 fragment from a plasmid, pBS/tlc1::LEU2 (a generous gift from F. Ishikawa), and selected on an SD plate lacking leucine. Diploid cells heterologous for tlc1::LEU2 were sporulated and dissected. Haploid cells with an appropriate genotype were selected and grown to saturation in YPDA medium at 30°. Every 24 hr, the cell density was measured by counting cells with a hemocytometer, and then the cultures were diluted with fresh YPDA medium to a density of 10⁵ cells/ml. This cycle was repeated for 10–16 days. At various time points, cells were collected for Southern blotting. Several independent isolates were analyzed for each mutant.

Single-colony streak assay:
Haploid cells were selected as described above and streaked on a YPDA plate. After incubation for 48 hr at 30°, single colonies were picked up and restreaked on a fresh YPDA plate. This restreaking was repeated 10 times to permit loss of viability and appearance of survivors. Single colonies from streak 7 were grown to saturation in YPDA medium, and cells were collected for DNA analysis.

Southern blotting of telomeres:
Genomic DNAs were digested with XhoI and separated in a 0.8% agarose gel. DNA was transferred onto a Hybond N membrane (Amersham, Buckinghamshire, UK) and UV crosslinked. The membrane was then hybridized with a ³²P-labeled, random-primed XhoI, EcoRI-fragment from a plasmid, pYNH3 (provided by F. Ishikawa). For characterization of the type II telomere pattern, DNAs were digested with AluI, HaeIII, HinfI, and MspI, subjected to electrophoresis in a 1.2% agarose gel, and then analyzed by Southern blotting. Blots were visualized using the phosphorimager BAS3000 (Fuji).

Physical analysis of mating-type switching:
Cells were grown to a density of 1 x 10⁷ cells/ml in YP-raffinose medium and HO endonuclease was induced by addition of 2% galactose. Strains were treated after 60 min, collected, and resuspended with YPDA medium to repress the GAL10::HO gene. At intervals, aliquots of cells were collected for DNA analysis. The DNA was digested with StyI and subjected to electrophoresis in a 0.8% agarose gel for 18 hr at 10 V/cm. DNA fragments were transferred onto a nylon membrane and hybridized with a ³²P-labeled probe. The probe was made using a PCR-amplified fragment: forward, 5'-TATGGCTATACCCTTATC, and reverse, 5'-GCATTTGAGTGGATACGC.

Determination of recombination frequencies:
Recombination frequencies were determined as described previously (RATTRAY and SYMINGTON 1994). Single colonies were grown on YPDA plates for 2–3 days. More than 15 pink colonies were resuspended in water and plated at the appropriate dilutions to determine total cell number and the number of Ade⁺ prototrophs. Median mitotic recombination frequencies were determined and rates (events/cell/generation) were calculated according to the following formula: rate = (0.4343 x median frequency)/(log N - log N₀), where N is the number of cells present in the colony and N₀ (number of initial cells) = 1.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

To examine the role of N- and C-terminal domains of Rad52 protein, we analyzed the effect of a C-terminal deletion mutant called rad52-329 on recombination. The rad52-329 allele, deleting 173 amino acids of a C-terminal Rad51-interacting domain of Rad52, encodes a hypothetical 296-amino-acid protein when it is translated from the third codon of the open reading frame (ADZUMA et al. 1984 ). Western blotting shows that this mutant cell expresses Rad52 protein with the expected size of 32 kD (data not shown). The mutant shows mild sensitivity to DNA-damaging agents (data not shown), as do other C-terminal deletion mutants of RAD52 such as rad52-327 (ADZUMA et al. 1984 ; BOUNDY-MILLS and LIVINGSTON 1993 ; ASLESON et al. 1999 ).

Repair of HO-induced DSBs is defective in a rad52-329 mutant:
To confirm that the rad52-329 mutant is indeed defective in RAD51-dependent recombination, we analyzed the mating-type switching, a gene conversion event completely dependent on RAD51 function (PAQUES and HABER 1999 ). Repair of an HO-induced DSB at the MAT locus is monitored at the DNA level after induction of HO endonuclease for 1 hr. To measure the formation of recombination products, the DNA samples were digested with StyI, which cut within Ya, but not Y sequences (Fig 2A). The appearance of a 1.0-kb StyI fragment is indicative of DSB repair from the HMRa locus. In the wild-type strain, the repair is efficient and completed within 150 min (Fig 2B). In the rad52-329 mutant, the DSB was introduced with similar kinetics as in wild type, but no recombinant molecule was formed. The DSB disappeared during further incubation, possibly due to the extensive degradation of the DSB ends. Consistent with this, we observed that the amount of a 2.3-kb DNA fragment next to the DSB fragment decreased in the mutant (lanes 19–22). These results indicate that the rad52-329 mutant is completely defective in the RAD51-dependent gene conversion at the MAT locus.

View larger version (60K): [in this window] [in a new window] [Download PPT slide]   Figure 2. Kinetics of mating-type switching. (A) A schematic representation of the MATa and MAT loci indicating the location of the StyI sites and the hybridization probe. (B) HO endonuclease produces a 0.7-kb fragment from the 1.9-kb MAT fragment. A 1.0-kb StyI fragment is produced when the mating type switches from MAT to MATa. Cells were exposed to galactose for 1 hr and incubated further in medium containing glucose. Time "0" is the time when the medium was exchanged. DNAs were isolated from cultures at indicated times and analyzed by Southern blotting. A nonspecific band from the rad52-329 allele is indicated by an asterisk. Wild type (YDB057), uninduced, lane 1; wild type, induced, lanes 2–11; rad52-329 (YMT546), uninduced, lane 12; rad52-329, induced, lanes 13–22.

rad52-329 reduces recombination between inverted repeats:
We also analyzed recombination between inverted repeats in the rad52-329 mutant. A previously described recombination substrate was used (RATTRAY and SYMINGTON 1994 ). The substrate is located on chromosome XV and consists of inverted heteroalleles of the ADE2 gene (Fig 3A). Both alleles are inactive, but the substrate can be rearranged by recombination events to form a functional ADE2 gene. Using this substrate, the rate of recombination was determined to be 3.78 x 10^-5 events/cell/generation in wild type and <4.2 x 10^-8 in a rad52 deletion mutant (Fig 3B), consistent with the previous report (RATTRAY and SYMINGTON 1994 ). The rad52-329 mutant reduces the rate fourfold relative to wild type. This rate is comparable to that in the rad51 mutant, but is much higher than that in the rad52 deletion. Thus, the rad52-329 mutant is partially defective in recombination between the repeats. Furthermore, we found that recombinants in the rad52-329 mutant were often accompanied with inversion of an intervening marker, TRP1, as seen in rad51 (data not shown). The inversion occurs in a nonconservative, RAD51-independent recombination (BAI et al. 1999 ; KANG and SYMINGTON 2000 ). Thus, the rad52-329 mutant is proficient in RAD51-independent recombination between the repeats.

View larger version (23K): [in this window] [in a new window] [Download PPT slide]   Figure 3. Recombination between inverted repeats. (A) A schematic representation of the recombination substrate (for more detail, see RATTRAY and SYMINGTON 1994 ). (B) Colonies from wild-type (yAR71), rad52-329 (YMT353), rad52 deletion (YMT362), and rad51 deletion (YMT367) cells were suspended in water and plated on YPDA and SD lacking adenine. Recombination rates were calculated as described in MATERIALS AND METHODS. At least 15 independent colonies were analyzed for each strain and the median values were used for the calculation.

In the absence of telomerase, the rad52-329 mutant initially shows senescence but later elongates its telomeres:
In the absence of the TLC1 gene, which encodes an RNA component of yeast telomerase, cells initially loose viability and later subsets of the cells survive by elongating their telomeres by recombination (LUNDBLAD and BLACKBURN 1993 ). We compared viability and telomere lengths of a set of isogenic mutants (in the background of S288C), including tlc1, tlc1 rad52, and tlc1 rad52-329. Cell viability was measured by serial streaking and serial dilution assays. In the serial dilution assay, cell viability was measured by diluting liquid cultures to 10⁵ cells/ml, allowing them to regrow for 24 hr, and counting the cell density (SINGER and GOTTSCHLING 1994 ; LE et al. 1999 ). Wild-type cells reach 10⁸ cells/ml in 24 hr, but tlc1 mutant cells that have lost viability or grow more slowly will reach only lower densities (see Fig 4). The ability to generate survivors was defined as the cells' ability to recover upon reaching their minimum growth rate after 6–9 days in culture. Since each cell seems to follow a unique fate in the telomere metabolism, we analyzed several independent cultures for each mutant, which were derived from independent isolates in the tetrad analysis of parental diploids.

View larger version (35K): [in this window] [in a new window] [Download PPT slide]   Figure 4. The rad52-329 and rad59 mutants elongate telomeres in the absence of telomerase. Cells were grown to saturation, counted every 24 hr with a hemocytometer, and then diluted to 105 cells/ml. The curves were the average of cultures from independent isolates with each genotype; "n" refers to the number of independent isolates analyzed. (A) Open circles, WT (n = 3); solid circles, tlc1 (n = 11); open triangles, tlc1 rad52-329 (n = 7); solid triangles, tlc1 rad52 (n = 20); open squares, tlc1 rad51 (n = 12); solid squares, tlc1 rad51 rad52-329 (n = 5). (B) Open circles, WT; solid circles, tlc1; open triangles, tlc1 rad59 fast-growing survivors (n = 6); solid triangles, tlc1 rad59 slow-growing survivors (n = 3). (C) Open circles, WT; solid circles, tlc1; open triangles, tlc1 rad59 fast-growing survivors; solid triangles, tlc1 rad51 rad59 (n = 3); open squares, tlc1 rad51; solid squares, tlc1 rad51 rad52-329 rad59 (n = 5). (D) Open circles, WT; solid circles, tlc1; open triangles, tlc1 rad52-329; solid triangles, tlc1 rad52-329 rad59 (n = 1); open squares, tlc1 rad59 slow-growing survivors.

The tlc1 mutant's growth rate in the S288C background declined gradually over 7 days and then increased as reported in other backgrounds (LUNDBLAD and BLACKBURN 1993 ; SINGER and GOTTSCHLING 1994 ). The tlc1 rad52-329 mutant's growth rate declined faster than that of the tlc1 mutant, although more slowly than that of the tlc1 rad52 mutant (Fig 4A). The tlc1 rad52 mutant produces few survivors (see below), but tlc1 rad52-329 generates survivors after 6 days.

We next constructed a tlc1 rad52-329 rad51 triple mutant and compared it to the tlc1 rad51 and tlc1 rad52-329 double mutants (Fig 4A). tlc1 rad51 shows an accelerated decline and generates survivors after 6 days as reported previously (LE et al. 1999 ; CHEN et al. 2001 ). tlc1 rad52-329 rad51 exhibits a survival curve similar to those of tlc1 rad51 and tlc1 rad52-329, suggesting that rad52-329 and rad51 mutant cells without telomerase expand their telomeres via the same pathway.

The rad52-329 mutant generates only type II survivors:
To examine which recombination pathway gives rise to survivors, telomere structures were analyzed. Genomic DNAs were digested with XhoI and analyzed by Southern blotting using a telomere probe (Fig 5). In the absence of telomerase, two distinct types of survivors are seen, as shown by the pattern of XhoI restriction fragments. In type I survivors, a short XhoI fragment <1 kb and strong amplified 5.2- and/or 6.7-kb Y' fragments are formed. Type II survivors have long tracts of telomeric TG_1–3 repeats and some amplified Y' fragments; the amplification of Y' in type II depends on the isolates examined (TENG and ZAKIAN 1999 ). In these type II survivors, the multiple distinct bands of telomeres between 1 and 6 kb or more (sometimes smeary) represent individual chromosomes with different lengths of TG_1–3 repeats. In the liquid assay, type II survivors grow faster than type I survivors and are expected to dominate (TENG et al. 2000 ). Consistent with this, tlc1 survivors predominantly showed the type II telomere pattern (Fig 5A). The tlc1 rad52-329 and tlc1 rad52-329 rad51 survivors also showed the type II telomere pattern, as did the survivors in the tlc1 rad51 mutant (Fig 5, B–D).

View larger version (83K): [in this window] [in a new window] [Download PPT slide]   Figure 5. The rad52-329 mutant generates type II survivors. An independent isolate of each mutant was grown in YPDA medium. On the days indicated at the top of each lane, genomic DNA was extracted and structures of telomeres were determined. Genomic DNA was digested with XhoI. Southern blotting was carried out as described in MATERIALS AND METHODS. Positions of long Y' (6.7 kb) and short Y' (5.2 kb) fragments are shown on the right. In the tlc1 mutant culture (A, lanes 4 and 5), some amplification of the short Y' fragment is seen in addition to the type II telomere pattern. Loss of the short Y' fragment was observed in fast-growing tlc1 rad59 and tlc1 rad51 rad59 rad52-329 mutant cultures, while the loss of long Y' was in some tlc1 cultures (E, lane 2). These losses of Y' fragments are dependent upon the isolates examined. (A) WT, lane 1; tlc1, lanes 2–5. (B) tlc1 rad52-329. (C) tlc1 rad51. (D) tlc1 rad51 rad52-329. (E) WT, lane 1; tlc1, lane 2; fast-growing tlc1 rad59 survivors, lanes 3–7; slow-growing tlc1 rad59 survivors, lanes 8–12. (F) WT, lane 1; tlc1, lane 2; tlc1 rad51 rad59 rad52-329, lanes 3–6.

Since liquid cultures select the fastest-growing cells, we also analyzed survivors from single colonies on plates. After survivors were generated, a single colony was picked and cultured overnight, and its genomic DNA was analyzed (Table 2). In 36 independent colonies of a tlc1 survivor, both type I and type II survivors were detected. The ratio of type I to type II is 0.9, which is less than that reported previously (TENG and ZAKIAN 1999 ; CHEN et al. 2001 ). On the other hand, the tlc1 rad52-329, tlc1 rad51, and tlc1 rad52-329 rad51 mutants give rise to type II survivors alone, even when analyzed by the plate assay. The fact that the rad52-329 mutant carries out type II telomere recombination shows that the C-terminal domain of Rad52 is dispensable for type II telomere recombination.

The tlc1 rad59 mutant generates both type I and type II survivors:
The structure of Rad52-329 is very similar to that of Rad59 (BAI and SYMINGTON 1996 ). We therefore examined the effect of a rad59 null mutation on telomere recombination in the absence of TLC1 and compared phenotypes to those of the tlc1 rad52-329 mutant. Previously, CHEN et al. 2001 reported that the tlc1 rad59 double mutant generates only type I survivors. However, in the liquid assay, we found that it can form both type I and type II survivors. Cultures from independent isolates of the tlc1 rad59 double mutant follow either of two fates (Fig 4B). On the basis of their growth curves and telomere structures, we refer to fast-growing and slow-growing tlc1 rad59 survivors (Fig 4B), which were also verified by Southern blotting to show type II and type I telomere patterns, respectively (Fig 5E). The fast-growing cells generated survivors after 8 days, while the slow-growing cells did so after 9 days. Moreover, the slow-growing survivors did not reach normal growth rate during subsequent culture. The presence of the type II telomere pattern in fast-growing tlc1 rad59 survivors was confirmed by digesting genomic DNA with four 4-base recognition restriction enzymes (Fig 6). This enabled us to measure the length of telomeric TG_1–3 repeats, since the enzymes digest most of the genome into small pieces, leaving the repeat intact (TENG et al. 2000 ). The fast-growing tlc1 rad59 survivors produced various lengths of longer TG_1–3 repeats, as seen in tlc1 cells. These results indicate that RAD59 is not essential for efficient type II telomere recombination.

View larger version (73K): [in this window] [in a new window] [Download PPT slide]   Figure 6. The tlc1 rad59 mutant generates both type I and type II survivors. Genomic DNA was prepared from each strain and digested with AluI, HaeIII, HinfI, and MspI. Southern blotting was performed as described in MATERIALS AND METHODS. The numbers at the top of the lanes represent the number of days that cells were grown. WT, lane 1; tlc1, lane 2; fast-growing tlc1 rad59 survivors, lanes 3–6; slow-growing tlc1 rad59 survivors, lanes 7–10.

In the liquid assay, 6 of 9 tlc1 rad59 cultures from independent isolates generate type II survivors, while the remaining 3 generate type I survivors (Table 3). However, the ratio might be an overestimate since the liquid assay favors fast-growing type II cells. Indeed, in the plate assay, only 2 of 10 independent tlc1 rad59 surviving colonies show the type II telomere pattern (Table 2). Therefore, the presence of the type II telomere pattern in tlc1 rad59 without growth selection confirms the above idea that the rad59 mutant without telomerase is proficient in type II recombination, but with a more reduced rate than that of the tlc1 strain.

We further confirmed that RAD59 is not essential for type II recombination by analyzing survivors in a tlc1 rad51 rad59 triple mutant, which eliminates the type I pathway. In the liquid assay, three of nine cultures from independent isolates of the tlc1 rad51 rad59 triple mutant generated survivors (Fig 4C), while the remaining six did not generate any survivors (Table 3). On the other hand, all isolates of the tlc1 rad51 and tlc1 rad59 double mutants produced survivors. Southern analysis confirmed the type II telomere pattern in the survivors (data now shown). These results are consistent with the hypothesis that RAD59 facilitates efficient RAD52-dependent type II recombination.

Our results on the presence of a type II telomere pattern in the tlc1 rad59 survivors are in contrast to those reported previously (CHEN et al. 2001 ). Since the strain backgrounds are different between previous and current studies, we reanalyzed a tlc1 rad59 strain used by Chen and colleagues. The tlc1 rad59 double mutant formed survivors slightly later than the tlc1 mutant (Figures 7A and 8), consistent with the previous results (CHEN et al. 2001 ). By analyzing a telomere structure in individual cultures, we found that 2 of 17 tlc1 rad59-independent isolates clearly showed a band pattern of type II (lane 1 in Fig 7B). The remaining 15 isolates exhibited a band pattern typical for type I survivors, but some of them contained faint bands of amplified TG_1–3 repeats (lanes 8 and 9 in Fig 7B). These results are consistent with the conclusion that Rad59 is necessary for most, but not all, type II recombination.

View larger version (41K): [in this window] [in a new window] [Download PPT slide]   Figure 7. The tlc1 rad59 mutant in a different strain background also generates type II survivors. JHUY564 was sporulated and survival of haploid colonies with a relevant genotype was analyzed in the liquid assay. The survival curves of each mutant (A) and Southern blots of telomere structure at 14 days in seven individual isolates (B) were obtained as described in MATERIALS AND METHODS. (A) Solid circles, tlc1; open triangles, tlc1 rad59 (an average of 14 independent isolates). (B) Telomere structure in tlc1 rad59 survivors, lanes 1–9. Some amplified TG1–3 repeats in type I survivors are indicated by arrowheads.

Rad52-329 alone promotes type II telomere recombination:
To examine whether the N-terminal portion of Rad52 alone has an ability to carry out recombination, we next analyzed the tlc1 rad52-329 rad59 triple mutant. In the liquid assay, only 1 of 11 independent isolates generated fast-growing survivors (Fig 4D), which were of type II (data not shown), while the remaining 10 did not generate any survivors. This indicates that the N-terminal domain of Rad52 can perform type II telomere recombination in the absence of RAD59, although the recombination in the mutant is very inefficient. However, a tlc1 rad51 rad52-329 rad59 quadruple mutant generates survivors more efficiently than the tlc1 rad52-329 rad59 triple mutant; five of seven cultures from independent isolates of the quadruple mutant gave rise to fast-growing survivors (Fig 4C and Table 3). The difference between the tlc1 rad52-329 rad59 triple mutant and the tlc1 rad51 rad52-329 rad59 quadruple mutant is statistically significant (P < 0.05, Student's t-test). All the survivors show the type II telomere pattern (Fig 5F). This supports the conclusion that Rad52-329 can recombine telomeres in the absence of Rad51 and Rad59. It also suggests that Rad51 inhibits type II telomere recombination.

The tlc1 rad52 double mutant generates few survivors (LUNDBLAD and BLACKBURN 1993 ): Of 25 independent isolates, 1 gave rise to survivors, and, on the basis of Southern blotting, these appeared to be type II (Table 3 and data not shown). However, these survivors are not typical type II, since they exhibit very slow growth, more like type I survivors. This suggests that even in the absence of RAD52, Rad51 may elongate the telomeres very inefficiently. Alternatively, it is possible to postulate a third minor telomere elongation pathway, which is independent of both RAD51 and RAD52. Since RAD51 inhibits RAD51-independent recombination, we reanalyzed survivors of a rad51 rad52 double mutant in the absence of TLC1. No survivors were recovered in the liquid assay (0/25).

The tlc1 rad52-329 rad50 mutant generates type II survivors:
Next, we analyzed the tlc1 rad52-329 rad50 triple mutant. RAD50 is required, but not essential for type II recombination (LE et al. 1999 ; CHEN et al. 2001 ). As shown previously, all independent isolates of the tlc1 rad50 double mutant generated slow-growing survivors with type I telomere pattern (data not shown). On the other hand, 2 of 14 isolates of the tlc1 rad52-329 rad50 triple mutant generated survivors, while the remaining 12 isolates did not generate any survivors (Fig 9A). The survivors in the triple mutant exhibited a typical type II telomere pattern (Fig 9B). Thus, we concluded that Rad52-329 protein can carry out type II telomere recombination even in the absence of RAD50 function.

View larger version (26K): [in this window] [in a new window] [Download PPT slide]   Figure 8. The tlc1 rad59 mutants in different strain backgrounds show different kinetics for generation of survivors. Cells were grown to saturation, counted every 24 hr with a hemocytometer, and then diluted to 105 cells/ml. The curves from eight independent isolates are shown. (A) tlc1 rad59 in S288C background. Open circles, triangles, and squares indicate fast-growing survivors; solid circles, triangles, and squares indicate slow-growing survivors. (B) tlc1 rad59 in JHUY564 background.

View larger version (41K): [in this window] [in a new window] [Download PPT slide]   Figure 9. The tlc1 rad52-329 rad50 mutant generates type II survivors. (A) The survival curves of each mutant. Solid circles, tlc1 (n = 11); open circles, tlc1 rad50 (n = 5); solid triangles, tlc1 rad50 rad52-329 (n = 2); "n" refers to the number of independent isolates analyzed. (B) Southern blots of telomere structure in tlc1 rad52-329 rad50 survivors showing the type II telomere pattern.

The tlc1 mre11 and tlc1 xrs2 mutants are similar to the tlc1 rad50 mutant:
Rad50 forms a complex with Mre11 and Rad50 (USUI et al. 1998 ). Previous reports showed some functional differences in telomere maintenance among components of the complex (BOULTON and JACKSON 1998 ; BUCHOLC et al. 2001 ). Different from the previous report (BOULTON and JACKSON 1998 ), all three mutant cells in the S288C background show a similar growth rate even at a high temperature such as 37° (data not shown). We also studied the effect of mre11 and xrs2 null mutations on telomere recombination (Fig 10). Both tlc1 mre11 and tlc1 xrs2 double mutants share the same phenotype as the tlc1 rad50 strain of generating survivors more slowly than the tlc1 mutant (Fig 10A). The mutants generate survivors more slowly than tlc1 alone. Most survivors from tlc1 mre11 or tlc1 xrs2 show the type I telomere pattern, but a few show the type II telomere pattern (Fig 10B and Table 3). Thus, all three members of the MRX complex are necessary for efficient type II recombination.

View larger version (49K): [in this window] [in a new window] [Download PPT slide]   Figure 10. tlc1 mre11 and tlc1 xrs2 are similar to tlc1 rad50. The survival curves of each mutant (A) and telomere structures (B) were obtained as described above. (A) Solid circles, tlc1 (n = 11); open circles, tlc1 rad50 (n = 7); open triangles, tlc1 mre11 (n = 14); solid triangles, tlc1 xrs2 (n = 14); "n" refers to the number of independent isolates analyzed. (B) tlc1 mre11 type I survivors, lanes 1–5; tlc1 mre11 type II survivors, lanes 6–10; tlc1 xrs2 type I survivors, lanes 11–15; tlc1 xrs2 type II survivors, lanes 16–20.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Previous genetic studies of yeast lacking telomerase have defined two recombination pathways (KASS-EISLER and GREIDER 2000 ). The formation of type I survivors requires the RAD51-dependent pathway, which involves RAD52, RAD54, RAD55, and RAD57 as well as RAD51. The formation of type II survivors requires the RAD51-independent pathway, which involves RAD50, RAD52, RAD59, MRE11, and XRS2. Recently, other factors that promote type II telomere extension have been identified: SGS1, MEC1, and TEL1 (COHEN and SINCLAIR 2001 ; HUANG et al. 2001 ; JOHNSON et al. 2001 ; TSAI et al. 2002 ). Among the genes involved in the RAD51-independent pathway, three components of the MRX complex are not essential for the formation of type II survivors (LE et al. 1999 ; CHEN et al. 2001 ). In contrast to the previous report (CHEN et al. 2001 ), we showed that the rad59 mutation reduces type II telomere recombination, but does not eliminate it. TID1/RDH54 is not necessary for telomere recombination (M. TSUKAMOTO and A. SHINOHARA, unpublished results). Thus, only RAD52 is an absolute requirement for type II telomere recombination among the RAD52 epistasis group.

One of the key questions in RAD51-independent recombination is to find which protein(s) is responsible for the search for homologous sequences and for strand invasion. It is proposed that type II survivors are generated through rolling-circle replication primed by an intrachromosomal telomeric D-loop (CHEN et al. 2001 ; NATARAJAN and MCEACHERN 2002 ). We speculate that Rad52 might catalyze the formation of the intrachromosomal D-loop. Since the N-terminal domain of Rad52 catalyzes annealing of complementary ssDNAs (MORTENSEN et al. 1996 ; SHINOHARA and OGAWA 1998 ; KAGAWA et al. 2001 ), this annealing activity might be involved in D-loop formation.

Our results described here indicate that Rad59 and the MRX complex are important, but not essential, for type II recombination. There might be multiple RAD51-independent recombination pathways at telomeres: RAD59 dependent and RAD59 independent or MRX dependent and MRX independent. Alternatively, there is one RAD51-independent recombination pathway in which Rad59 and the MRX complex facilitate the recombination catalyzed by Rad52. The genetic requirement for telomere recombination supports the view that type II telomere recombination is mechanistically similar to BIR. However, recombination at telomeres is slightly different from BIR events at other genomic loci. BIR initiated by DSB at the MAT locus requires a specific cis-acting sequence (MALKOVA et al. 2001 ). Chromosome ends might contain such an element that facilitates BIR. Alternatively, a constraint, which enforces the requirement for the BIR facilitator on internal chromosomal loci, might be alleviated at the end of chromosomes.

Since telomeres are composed of irregular TG_1–3 repeats (SHAMPAY et al. 1984 ), there is a high potential for mismatches and short homology between terminal sequences when the sequences recombine. Thus, it is thought that RAD51-independent telomere recombination is homeologous BIR, which depends on a short homology. Interestingly, deletion of a mismatch repair gene, MSH2, alleviates the growth defect of telomerase-deficient cells, suggesting that homeologous recombination is suppressed in cells without telomerase (RIZKI and LUNDBLAD 2001 ). On the other hand, longer-sequence homology is necessary for type I recombination. Consistent with this idea, Haber and his colleagues showed that a minimum homology length required for the RAD51-independent recombination is 30 bp, while that for the RAD51-dependent recombination is 100 bp (IRA and HABER 2002 ).

RAD51-dependent and RAD51-independent telomere recombination pathways seem to be partially exclusive of each other. Amplification of Y' elements is rarely seen in type II survivors, while amplification of TG_1–3 repeats is not observed in type I survivors of the tlc1. It is possible to consider that a factor, which determines the choice of telomere recombination pathways, might fluctuate in each tlc1 isolate (or cell). Such a factor might be a positive and/or negative regulator for the recombination pathways. Our results that Rad51 has a positive effect on type I, but a negative one on type II recombination, suggest that Rad51 might be a protein whose amounts in a cell determine the pathway choice (see below). Furthermore, it is shown that Cdc13, a telomere end-binding protein, or the Ku complex suppresses RAD51-dependent telomere recombination (GRANDIN and CHARBONNEAU 2003 ), suggesting that the composition of telomere ends determines a cell's ability to choose a telomere recombination pathway. We showed that the tlc1 rad59 mutants in different strain backgrounds exhibit similar, but distinct, phenotypes in the choice of recombination pathways to elongate telomeres. These results suggest that a genetic difference between strains and/or even in a cell determines which recombination pathway is predominantly used to elongate telomeres in the absence of telomerase.

Telomere elongation by recombination is considered to be a rare event. However, given that all 32 ends of chromosomes should be elongated in survivors of tlc1 cells, we might underestimate the rate of recombination per chromosome end. Further analysis is necessary to determine how frequent telomere recombination is. Interestingly, in some insects such as the mosquito Anopheles and the dipteran Chironomus, telomere recombination is thought to be the sole mechanism for maintaining the repeats at chromosome ends (LOPEZ et al. 1996 ; ROTH et al. 1997 ). At least in these organisms, recombination seems to be an efficient process for telomere elongation.

	FOOTNOTES

¹ Present address: Research Institute for Radiation Biology and Medicine, Hiroshima University, Hiroshima 734-8553, Japan.

	ACKNOWLEDGMENTS

We are grateful to Akiyo Yamazaki and Harumi Kato for construction of strains and plasmids; to Hisao Masukata, Takuro Nakagawa, and Neil Hunter for helpful discussion; to Lorraine Symington for unpublished results; and to Jim Haber, Fuyuki Ishikawa, Tomoko Ogawa, and Rodney Rothstein for plasmids and Carol Greider and Lorraine Symington for strains. This work was supported by grants from the Ministry of Education, Science and Culture of Japan to Priority Area to A.S.

Manuscript received April 21, 2003; Accepted for publication August 19, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ADZUMA, K., T. OGAWA, and H. OGAWA, 1984  Primary structure of the RAD52 gene in Saccharomyces cerevisiae.. Mol. Cell. Biol. 4:2735-2744.[Abstract/Free Full Text]

ALANI, E., L. CAO, and N. KLECKNER, 1987  A method for gene disruption that allows repeated use of URA3 selection in the construction of multiply disrupted yeast strains. Genetics 116:541-545.[Abstract/Free Full Text]

ALANI, E., S. SUBBIAH, and N. KLECKNER, 1989  The yeast RAD50 gene encodes a predicted 153-kD protein containing a purine nucleotide-binding domain and two large heptad-repeat regions. Genetics 122:47-57.[Abstract/Free Full Text]

ASLESON, E. N., R. J. OKAGAKI, and D. M. LIVINGSTON, 1999  A core activity associated with the N terminus of the yeast RAD52 protein is revealed by RAD51 overexpression suppression of C-terminal rad52 truncation alleles. Genetics 153:681-692.[Abstract/Free Full Text]

BAI, Y. and L. S. SYMINGTON, 1996  A Rad52 homolog is required for RAD51-independent mitotic recombination in Saccharomyces cerevisiae.. Genes Dev. 10:2025-2037.[Abstract/Free Full Text]

BAI, Y., A. P. DAVIS, and L. S. SYMINGTON, 1999  A novel allele of RAD52 that causes severe DNA repair and recombination deficiencies only in the absence of RAD51 or RAD59.. Genetics 153:1117-1130.[Abstract/Free Full Text]

BOULTON, S. J. and S. P. JACKSON, 1998  Components of the Ku-dependent non-homologous end-joining pathway are involved in telomeric length maintenance and telomeric silencing. EMBO J. 17:1819-1828.[Medline]

BOUNDY-MILLS, K. L. and D. M. LIVINGSTON, 1993  A Saccharomyces cerevisiae RAD52 allele expressing a C-terminal truncation protein: activities and intragenic complementation of missense mutations. Genetics 133:39-49.[Abstract]

BUCHOLC, M., Y. PARK, and A. J. LUSTIG, 2001  Intrachromatid excision of telomeric DNA as a mechanism for telomere size control in Saccharomyces cerevisiae.. Mol. Cell. Biol. 21:6559-6573.[Abstract/Free Full Text]

CHEN, Q., A. IJPMA, and C. W. GREIDER, 2001  Two survivor pathways that allow growth in the absence of telomerase are generated by distinct telomere recombination events. Mol. Cell. Biol. 21:1819-1827.[Abstract/Free Full Text]

COHEN, H. and D. A. SINCLAIR, 2001  Recombination-mediated lengthening of terminal telomeric repeats requires the Sgs1 DNA helicase. Proc. Natl. Acad. Sci. USA 98:3174-3179.[Abstract/Free Full Text]

DAVIS, A. P. and L. S. SYMINGTON, 2001  The yeast recombinational repair protein Rad59 interacts with Rad52 and stimulates single-strand annealing. Genetics 159:515-525.[Abstract/Free Full Text]

GIETZ, R. D. and A. SUGINO, 1988  New yeast-Escherichia coli shuttle vectors constructed with in vitro mutagenized yeast genes lacking six-base pair restriction sites. Gene 74:527-534.[Medline]

GRANDIN, N. and M. CHARBONNEAU, 2003  The Rad51 pathway of telomerase-independent maintenance of telomeres can amplify TG_1–3 sequences in yku and cdc13 mutants of Saccharomyces cerevisiae.. Mol. Cell. Biol. 23:3721-3734.[Abstract/Free Full Text]

HUANG, P., F. E. PRYDE, D. LESTER, R. L. MADDISON, and R. H. BORTS et al., 2001  SGS1 is required for telomere elongation in the absence of telomerase. Curr. Biol. 11:125-129.[Medline]

IRA, G. and J. E. HABER, 2002  Characterization of RAD51-independent break-induced replication that acts preferentially with short homologous sequences. Mol. Cell. Biol. 22:6384-6392.[Abstract/Free Full Text]

JOHNSON, F. B., R. A. MARCINIAK, M. MCVEY, S. A. STEWART, and W. C. HAHN et al., 2001  The Saccharomyces cerevisiae WRN homolog Sgs1p participates in telomere maintenance in cells lacking telomerase. EMBO J. 20:905-913.[Medline]

JOHZUKA, K. and H. OGAWA, 1995  Interaction of Mre11 and Rad50: two proteins required for DNA repair and meiosis-specific double-strand break formation in Saccharomyces cerevisiae.. Genetics 139:1521-1532.[Abstract]

KAGAWA, W., H. KURUMIZAKA, S. IKAWA, S. YOKOYAMA, and T. SHIBATA, 2001  Homologous pairing promoted by the human Rad52 protein. J. Biol. Chem. 276:35201-35208.[Abstract/Free Full Text]

KAGAWA, W., H. KURUMIZAKA, R. ISHITANI, S. FUKAI, and O. NUREKI et al., 2002  Crystal structure of the homologous-pairing domain from the human Rad52 recombinase in the undecameric form. Mol. Cell 10:359-371.[Medline]

KANG, L. E. and L. S. SYMINGTON, 2000  Aberrant double-strand break repair in rad51 mutants of Saccharomyces cerevisiae.. Mol. Cell. Biol. 20:9162-9172.[Abstract/Free Full Text]

KASS-EISLER, A. and C. W. GREIDER, 2000  Recombination in telomere-length maintenance. Trends Biochem. Sci. 25:200-204.[Medline]

KRAUS, E., W. Y. LEUNG, and J. E. HABER, 2001  Break-induced replication: a review and an example in budding yeast. Proc. Natl. Acad. Sci. USA 98:8255-8262.[Abstract/Free Full Text]

LE, S., J. K. MOORE, J. E. HABER, and C. W. GREIDER, 1999  RAD50 and RAD51 define two pathways that collaborate to maintain telomeres in the absence of telomerase. Genetics 152:143-152.[Abstract/Free Full Text]

LOPEZ, C. C., L. NIELSEN, and J. E. EDSTROM, 1996  Terminal long tandem repeats in chromosomes from Chironomus pallidivittatus.. Mol. Cell. Biol. 16:3285-3290.[Abstract]

LUNDBLAD, V. and E. H. BLACKBURN, 1993  An alternative pathway for yeast telomere maintenance rescues est1-senescence. Cell 73:347-360.[Medline]

MALKOVA, A., E. L. IVANOV, and J. E. HABER, 1996  Double-strand break repair in the absence of RAD51 in yeast: a possible role for break-induced DNA replication. Proc. Natl. Acad. Sci. USA 93:7131-7136.[Abstract/Free Full Text]

MALKOVA, A., L. SIGNON, C. B. SCHAEFER, M. L. NAYLOR, and J. F. THEIS et al., 2001  RAD51-independent break-induced replication to repair a broken chromosome depends on a distant enhancer site. Genes Dev. 15:1055-1060.[Abstract/Free Full Text]

MCEACHERN, M. J., A. KRAUSKOPF, and E. H. BLACKBURN, 2000  Telomeres and their control. Annu. Rev. Genet. 34:331-358.[Medline]

MILNE, G. T. and D. T. WEAVER, 1993  Dominant negative alleles of RAD52 reveal a DNA repair/recombination complex including Rad51 and Rad52. Genes Dev. 7:1755-1765.[Abstract/Free Full Text]

MORROW, D. M., C. CONNELLY, and P. HIETER, 1997  "Break copy" duplication: a model for chromosome fragment formation in Saccharomyces cerevisiae.. Genetics 147:371-382.[Abstract]

MORTENSEN, U. H., C. BENDIXEN, I. SUNJEVARIC, and R. ROTHSTEIN, 1996  DNA strand annealing is promoted by the yeast Rad52 protein. Proc. Natl. Acad. Sci. USA 93:10729-10734.[Abstract/Free Full Text]

MURIS, D. F., O. BEZZUBOVA, J. M. BUERSTEDDE, K. VREEKEN, and A. S. BALAJEE 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.[Medline]

NATARAJAN, S. and M. J. MCEACHERN, 2002  Recombinational telomere elongation promoted by DNA circles. Mol. Cell. Biol. 22:4512-4521.[Abstract/Free Full Text]

NEW, J. H., T. SUGIYAMA, E. ZAITSEVA, and S. C. KOWALCZYKOWSKI, 1998  Rad52 protein stimulates DNA strand exchange by Rad51 and replication protein A. Nature 391:407-410.[Medline]

PÂQUES, F. and J. E. HABER, 1999  Multiple pathways of recombination induced by double-strand breaks in Saccharomyces cerevisiae.. Microbiol. Mol. Biol. Rev. 63:349-404.[Abstract/Free Full Text]

PARSONS, C. A., P. BAUMANN, E. VAN DYCK, and S. C. WEST, 2000  Precise binding of single-stranded DNA termini by human RAD52 protein. EMBO J. 19:4175-4181.[Medline]

RATTRAY, A. J. and L. S. SYMINGTON, 1994  Use of a chromosomal inverted repeat to demonstrate that the RAD51 and RAD52 genes of Saccharomyces cerevisiae have different roles in mitotic recombination. Genetics 138:587-595.[Abstract]

REID, R. J., M. LISBY, and R. ROTHSTEIN, 2002  Cloning-free genome alterations in Saccharomyces cerevisiae using adaptamer-mediated PCR. Methods Enzymol. 350:258-277.[Medline]

RIZKI, A. and V. LUNDBLAD, 2001  Defects in mismatch repair promote telomerase-independent proliferation. Nature 411:713-716.[Medline]

ROTH, C. W., F. KOBESKI, M. F. WALTER, and H. BIESSMANN, 1997  Chromosome end elongation by recombination in the mosquito Anopheles gambiae.. Mol. Cell. Biol. 17:5176-5183.[Abstract]

SHAMPAY, J., J. W. SZOSTAK, and E. H. BLACKBURN, 1984  DNA sequences of telomeres maintained in yeast. Nature 310:154-157.[Medline]

SHINOHARA, A. and T. OGAWA, 1998  Stimulation by Rad52 of yeast Rad51-mediated recombination. Nature 391:404-407.[Medline]

SHINOHARA, A., H. OGAWA, and T. OGAWA, 1992  Rad51 protein involved in repair and recombination in S. cerevisiae is a RecA-like protein. Cell 69:457-470.[Medline]

SHINOHARA, A., M. SHINOHARA, T. OHTA, S. MATSUDA, and T. OGAWA, 1998  Rad52 forms ring structures and co-operates with RPA in single-strand DNA annealing. Genes Cells 3:145-156.[Abstract]

SIGNON, L., A. MALKOVA, M. L. NAYLOR, H. KLEIN, and J. E. HABER, 2001  Genetic requirements for RAD51- and RAD54-independent break-induced replication repair of a chromosomal double-strand break. Mol. Cell. Biol. 21:2048-2056.[Abstract/Free Full Text]

SINGER, M. S. and D. E. GOTTSCHLING, 1994  TLC1: template RNA component of Saccharomyces cerevisiae telomerase. Science 266:404-409.[Abstract/Free Full Text]

SINGLETON, M. R., L. M. WENTZELL, Y. LIU, S. C. WEST, and D. B. WIGLEY, 2002  Structure of the single-strand annealing domain of human RAD52 protein. Proc. Natl. Acad. Sci. USA 99:13492-13497.[Abstract/Free Full Text]

STASIAK, A. Z., E. LARQUET, A. STASIAK, S. MULLER, and A. ENGEL et al., 2000  The human Rad52 protein exists as a heptameric ring. Curr. Biol. 10:337-340.[Medline]

SUGIYAMA, T., J. H. NEW, and S. C. KOWALCZYKOWSKI, 1998  DNA annealing by RAD52 protein is stimulated by specific interaction with the complex of replication protein A and single-stranded DNA. Proc. Natl. Acad. Sci. USA 95:6049-6054.[Abstract/Free Full Text]

SUNG, P., 1994  Catalysis of ATP-dependent homologous DNA pairing and strand exchange by yeast RAD51 protein. Science 265:1241-1243.[Abstract/Free Full Text]

SUNG, P., 1997  Function of yeast Rad52 protein as a mediator between replication protein A and the Rad51 recombinase. J. Biol. Chem. 272:28194-28197.[Abstract/Free Full Text]

SYMINGTON, L. S., 2002  Role of RAD52 epistasis group genes in homologous recombination and double-strand break repair. Microbiol. Mol. Biol. Rev. 66:630-670.[Abstract/Free Full Text]

TENG, S. C. and V. A. ZAKIAN, 1999  Telomere-telomere recombination is an efficient bypass pathway for telomere maintenance in Saccharomyces cerevisiae.. Mol. Cell. Biol. 19:8083-8093.[Abstract/Free Full Text]

TENG, S. C., J. CHANG, B. MCCOWAN, and V. A. ZAKIAN, 2000  Telomerase-independent lengthening of yeast telomeres occurs by an abrupt Rad50p-dependent, Rif-inhibited recombinational process. Mol. Cell 6:947-952.[Medline]

TSAI, Y. L., S. F. TSENG, S. H. CHANG, C. C. LIN, and S. C. TENG, 2002  Involvement of replicative polymerases, Tel1p, Mec1p, Cdc13p, and the Ku complex in telomere-telomere recombination. Mol. Cell. Biol. 22:5679-5687.[Abstract/Free Full Text]

USUI, T., T. OHTA, H. OSHIUMI, J. TOMIZAW