Efficient Repair of All Types of Single-Base Mismatches in Recombination Intermediates in Chinese Hamster Ovary Cells: Competition Between Long-Patch and G-T Glycosylase-Mediated Repair of G-T Mismatches

Corresponding author: Jac A. Nickoloff, Department of Molecular Genetics and Microbiology, School of Medicine, University of New Mexico, Albuquerque, NM 87131., jnickoloff{at}salud.unm.edu (E-mail).

Communicating editor: L. S. SYMINGTON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Repair of all 12 single-base mismatches in recombination intermediates was investigated in Chinese hamster ovary cells. Extrachromosomal recombination was stimulated by double-strand breaks in regions of shared homology. Recombination was predicted to occur via single-strand annealing, yielding heteroduplex DNA (hDNA) with a single mismatch. Nicks were expected on opposite strands flanking hDNA, equidistant from the mismatch. Unlike studies of covalently closed artificial hDNA substrates, all mismatches were efficiently repaired, consistent with a nick-driven repair process. The average repair efficiency for all mispairs was 92%, with no significant differences among mispairs. There was significant strand-independent repair of G-T G-C, with a slightly greater bias in a CpG context. Repair of C-A was also biased (toward C-G), but no A-C G-C bias was found, a possible sequence context effect. No other mismatches showed evidence of biased repair, but among hetero-mismatches, the trend was toward retention of C or G vs. A or T. Repair of both T-T and G-T mismatches was much less efficient in mismatch repair-deficient cells (~25%), and the residual G-T repair was completely biased toward G-C. Our data indicate that single-base mismatches in recombination intermediates are substrates for at least two competing repair systems.

BASE pair mismatches arise constantly in genomic DNA as a consequence of metabolic processes. Mismatches can result from incorrect base insertion, slippage of DNA polymerase during replication, or strand exchange during homologous recombination. The most frequent mismatch arises from spontaneous or genotoxin-induced deamination of either 5-methylcytosine, which accounts for 2–8% of all cytosine moieties (DOEFLER 1983 ; RIGGS and JONES 1983 ), or cytosine to form G-T and G-U mismatches, respectively. The frequency of these deamination processes is far greater than the observed transversion mutations expected for random repair of G-T or G-U mismatches (SHEN et al. 1994 ) consistent with mechanisms that restore these to G-C (BROWN and JIRICNY 1987 ; SCHMUTTE et al. 1995 ).

Mismatch repair is necessary to reduce the accumulation of mutations and to maintain genome integrity. In bacteria and yeast, mismatch repair-deficient cells show genome instability and a mutator phenotype (LEVINSON and GUTMAN 1987 ; STRAND et al. 1993 ), and in humans, mutations in mismatch repair genes are linked to the genesis of certain cancers, including hereditary nonpolyposis colon cancer (BRONNER et al. 1994 ; BOYER et al. 1995 ; PROLLA et al. 1998 ). In recent years, it has become apparent that similar mismatch repair mechanisms operate in both prokaryotes and eukaryotes (reviewed by MODRICH 1991 ). There are three distinct mismatch repair mechanisms in Escherichia coli. One is long-patch repair mediated by mutHLS, which can recognize single or multiple mismatches and excise a stretch, often >1 kb, of newly synthesized DNA (reviewed by RASMUSSEN et al. 1998 ). In addition, there are two short-patch repair pathways, involving vsr and mutY, that mediate G-T G-C and G-A G-C repair, respectively (LIEB 1983 ; NGHIEM et al. 1988 ; SOHAIL et al. 1990 ; TSAI-WU et al. 1991 ). Although less well-characterized than in bacteria, a long-patch repair system has been suggested for Saccharomyces cerevisiae and mammalian cells because these have homologues of mutS and mutL (MSH and MLH) (KRAMER et al. 1989B ; REENAN and KOLODNER 1992 ). There is also evidence for short-tract mismatch repair systems in eukaryotes. A thymine DNA glycosylase specific for G-T mismatches was identified in HeLa cell extracts. This enzyme removes T in G-T mispairs, correcting to G-C (WIEBAUER and JIRICNY 1989 , WIEBAUER and JIRICNY 1990 ). A mammalian homologue of mutY (MYH) also has been characterized (MCGOLDRICK et al. 1995 ).

In vertebrate cells the repair of all possible mispaired bases has been studied using either cell extracts (BROOKS et al. 1989 ; HOLMES et al. 1990 ; VARLET et al. 1990 ; THOMAS et al. 1991 ) or in vivo by transfection of DNA into cells containing mismatches created in vitro (FOLGER et al. 1985 ; BROWN and JIRICNY 1988 ; HEYWOOD and BURKE 1990 ). Although these studies indicated that specific mismatch repair can occur, there was a large variability in the overall efficiency and direction of repair among different mismatch types. Several factors may influence the direction and efficiency of mismatch repair. For example, an early study suggested that methylation status and DNA nicks were important determinants of mismatch repair strand discrimination in simian kidney cells (HARE and TAYLOR 1985 ). Many studies have since shown that DNA nicks increase the frequency of mismatch repair and bias repair toward the strand containing the nick (HOLMES et al. 1990 ; THOMAS et al. 1991 ; UMAR et al. 1994 ; MILLER et al. 1997 ; TAGHIAN and NICKOLOFF 1998 ). However, nick-directed repair may not be general; HEYWOOD and BURKE 1990 reported that nicks do not drive the direction of mismatch repair in simian COS-7 cells. Mismatch repair efficiency of particular mismatches is also influenced by the presence of other nearby mismatches, thought to reflect co-repair. Thus, a poorly repaired mismatch can be repaired efficiently when located near a well-repaired mismatch (PETES et al. 1991 ; CARRAWAY and MARINUS 1993 ; WENG and NICKOLOFF 1998 ).

In a previous study we investigated mismatch repair in heteroduplex DNA (hDNA) formed by extrachromosomal recombination between plasmid substrates in Chinese hamster ovary (CHO) cells (DENG and NICKOLOFF 1994 ). hDNA was predicted to be flanked by nicks on opposite strands. We found that two single-base mismatches, located in close proximity, were repaired with strong bias toward loss of the mismatched bases on the strand nearest to one of the flanking nicks. In two subsequent investigations with substrates that produced three to nine mismatches, all mismatch types tested were efficiently repaired in CHO cells, and product spectra were consistent with most repair being directed from nicks (MILLER et al. 1997 ; TAGHIAN and NICKOLOFF 1998 ). Because these substrates produced hDNA with multiple mismatches, the repair of each single-base mismatch was potentially influenced by the asymmetry of mismatch locations relative to flanking nicks and/or by co-repair of neighboring mismatches. The present study with individual mismatches located equidistant from flanking nicks was designed to investigate the repair of all types of single-base mismatches in the absence of these confounding factors. We report that all single-base mismatches are repaired with high efficiency. The majority of mismatches showed no repair bias, consistent with a nick-driven repair process. However, the repair of G-T mismatches was biased toward loss of T, and this bias was strand-independent. Crosses were also performed in mutant cells deficient in mismatch repair ("clone B"; BRANCH et al. 1993 ; TAGHIAN and NICKOLOFF 1998 ). Clone B cells share several characteristics with mutants defective in mismatch repair, such as tolerance to alkylating agents, increased spontaneous mutagenesis and microsatellite instability, defects in mismatch binding (KAT et al. 1993 ; AQUILINA et al. 1994 ; HESS et al. 1994 ), and increased mismatch segregation rates (TAGHIAN and NICKOLOFF 1998 ). Clone B cell extracts fail to complement extracts of LoVo cells, which are defective in both alleles of hMSH2, and it was suggested that clone B and LoVo share this defect (AQUILINA et al. 1994 ). Msh2-deficient mice also exhibit microsatellite instability, a mutator phenotype, and methylation tolerance (DE WIND et al. 1995 ). We show that repair of two mismatches (T-T and G-T) is greatly reduced in clone B cells, but residual repair of G-T still favored loss of T. These results indicate that single-base mismatches in recombination intermediates are subject to repair by at least two distinct and competing mechanisms.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Plasmid DNA constructions:
Plasmids were constructed by standard procedures (SAMBROOK et al. 1989 ). For transfection into CHO cells, highly supercoiled plasmid DNA was isolated by acidic-phenol extraction (WANG and ROSSMAN 1994 ). pneoAn was created by inserting the 2.1-kbp HindIII-BamHI fragment of pSV2neo into the HindIII and BamHI sites of pUC19. Three single-base substitutions were created at position 430 of the neo gene in pneoAn (DENG and NICKOLOFF 1994 ) and pSV2neo (SOUTHERN and BERG 1982 ) by unique-site elimination mutagenesis (DENG and NICKOLOFF 1994 ). A frameshift mutation was introduced into the resulting pSV2neo derivatives by insertion of a 10-bp XhoI linker (5'-CCCTCGAGGG-3') into the BssHII site of neo to create four pSV2neoX(B) derivatives. Eight plasmids were used as recombination substrates, with four plasmids each derived from pSV2neoX(B) and pneoAn containing each of the four possible bases at position 430 in neo. An ApaI site was created when the coding strand carried a C at position 430.

Cell culture and recombination assays:
Wild-type (K1c) and mismatch-repair mutant (clone B) CHO cells were cultured as described previously (MILLER et al. 1997 ; TAGHIAN and NICKOLOFF 1998 ). For transfection experiments, derivatives of pSV2neoX(B) and pneoAn were linearized with MscI and BglII, respectively, and purified through Sepharose CL-6B (NICKOLOFF 1994 ). Electrotransformations of CHO cells were performed essentially as described by TAGHIAN and NICKOLOFF 1995 , using the following conditions: 200 ng of each linearized plasmid was electroporated into a suspension of 6 x 10⁶ cells in 0.75 ml of PBS, using a pulse of 300 V, 960 µF and a 0.4-cm electrode gap. Plating efficiency for electroporated cells averaged 65% that of untreated controls. G418^r recombinants were selected, and genomic DNA was prepared and used for plasmid rescue or PCR amplification as described (TAGHIAN and NICKOLOFF 1998 ). Restriction mapping of rescued plasmids and PCR products readily distinguished recombinants from nonrecombinants (TAGHIAN and NICKOLOFF 1998 ).

The direction of repair of single-base mismatches was determined either by digestion with ApaI or by a modification of an oligonucleotide hybridization procedure (NICKOLOFF and HALLICK 1982 ). Briefly, rescued plasmids or PCR products were electrophoresed through duplicate 0.8% agarose gels and transferred to Nytran membranes (Schleicher and Schuell, Keene, NH). Membranes were prehybridized for 1 hr in Church's buffer (0.5 M NaPO₄, pH 7.2, 1 mM EDTA, pH 8.0, 7% SDS; CHURCH and GILBERT 1984 ) containing 100 µg/ml denatured salmon sperm DNA, and hybridized for 2 hr in the same solution containing ³²P-end-labeled DNA probe at a temperature 5° below the T_m of the probe. For each sample, two hybridizations were performed with 15-base oligonucleotide probes (5'-TCTGTTGNGCCCAGT-3', where N = A, C, G, or T) complementary to each of the two possible repair outcomes of particular single-base mismatches. Membranes were washed in 2x SSC (0.15 M NaCl, 0.015 M sodium citrate), 0.1% SDS, for 20 min each at 37, 42, and 49° and exposed to Kodak X-Omat AR film. In each case, one probe was completely complementary to the target and the other had a mismatch at the central base; this mismatch destabilizes the hybrid and causes the probe to be washed off at lower stringency. By performing parallel hybridizations, we obtained positive and negative signals for each sample tested. Mismatch-repair efficiencies were measured using PCR-amplified DNA from products with single, integrated recombinant molecules. Repaired recombinants hybridized to only one probe. Products that escaped repair segregated at the first round of replication after integration (producing a sectored colony) and hybridized to both probes. Repair data were analyzed by the StatXact-3 program (Statistical Solutions Limited) using a two-sided binomial distribution at the 95% confidence level.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Experimental design:
The repair of single-base mismatches formed in vivo in hDNA intermediates by extrachromosomal recombination was studied with two types of plasmid recombination substrates as described previously (DENG and NICKOLOFF 1994 ; MILLER et al. 1997 ; TAGHIAN and NICKOLOFF 1998 ). pSV2neo derivatives contain a neo gene regulated by an SV40 promoter and inactivated by a linker frameshift mutation. pneoAn derivatives contain an inactive neo gene because they lack a promoter (Figure 1). An SV40-driven neo⁺ gene formed by recombination can integrate and produce G418^r transfectants. Although extrachromosomal recombination can occur by more than one mechanism; (DESAUTELS et al. 1990 ; YANG and WALDMAN 1992 ; see DISCUSSION), single-strand annealing (SSA) (LIN et al. 1987 , LIN et al. 1990 ; CARROLL et al. 1994 ) is considered to be the primary recombination pathway. SSA might involve annealing between single strands exposed by either a 5' to 3' exonuclease or a 3' to 5' exonuclease. Although an early report suggested 3' to 5' exonuclease activity in mouse L cells (LIN et al. 1987 ), recent data indicate that ends are processed by a 5' to 3' exonuclease in both mouse ES cells and yeast (SUN et al. 1991 ; HUANG and SYMINGTON 1993 ; HENDERSON and SIMONS 1997 ). Below we describe genetic evidence for 5' to 3' exonuclease activity mediating SSA in CHO cells.

View larger version (30K): In this window In a new window Download PPT slide   Figure 1. Structures of recombination substrates and hDNA intermediates. pSV2neo and pneoAn derivatives were cleaved with MscI and BglII, respectively (top). Following transfection, these ends are processed to single-stranded regions, exposing complementary strands (middle, left) that anneal to produce an hDNA intermediate with single-strand breaks in opposite strands (triangles) corresponding to DSB sites (middle, right). The diagram shows the formation of the G-T mismatch. Following integration, recombinant molecules are rescued from genomic DNA by digesting with EcoRI, recircularizing with T4 DNA ligase, and transforming E. coli. Below are shown sequence contexts around position 430 in neo. In the examples shown, the G-T mismatch mimics deamination of CpG to TpG (boxed), whereas the T-G mismatch mimics deamination of CpC to TpC (boxed).

Key predictions of the SSA model are that hDNA in recombination intermediates may be flanked by nicks, located on opposite strands corresponding to the restriction cut sites (Figure 1), and that a nonconservative crossover product results. Depending on the extent of exonuclease digestion, SSA intermediates may include more or less hDNA in the region between the double-strand breaks (DSBs), with maximum hDNA (242 bp) formed if exonuclease digestion extends from broken ends to (or past) DSB positions in recombining partners. Lesser digestion would produce less hDNA and annealed plasmids would contain flaps with 3' ends; such flaps could be removed by homologs of yeast Rad1p/10p and human ERCC1/4 (IVANOV and HABER 1995 ; BROOKMAN et al. 1996 ). Greater digestion would produce maximum hDNA flanked by single-stranded gaps that could be filled by extension of 3' ends. By using this model to predict structures of recombination intermediates, we performed 12 crosses to produce all types of single-base mismatches at position 430 of the neo gene. Because position 430 is essentially equidistant from the two initiating breaks, markers at this position are expected to occur in hDNA at maximum rates. Data presented below indicate that mismatches were produced at position 430 at least 75% of the time. Mismatch repair may occur before or after integration of the extrachromosomal DNA into the genome. If a mismatch is not repaired, the resultant sectored colony can be identified by segregation analysis. Repair was analyzed in wild-type (K1c) and mismatch repair-deficient mutant (clone B; BRANCH et al. 1993 ) CHO cells.

Silence of markers and recombination frequencies:
The point mutations introduced into position 430 of neo were at a codon third position and were expected to maintain the neo amino acid sequence. The silence of these mutations was confirmed by electroporating pSV2neo derivatives into CHO cells and selecting for G418-resistant colonies, and by transformation of E. coli with derivatives of pSV2neo or pneoAn and selecting for kan^r colonies. Each of the six mutant plasmids and the two wild-type plasmids reproducibly yielded similar numbers of G418^r or kan^r colonies (data not shown). Recombination frequencies in CHO K1c cells were 5–10 x 10^-6/0.1 µg of plasmid DNA for each of 12 crosses (Table 1). Recombination frequencies for G-T and T-T crosses, measured in parallel in K1c and clone B cells to control for variation in the input plasmid DNAs, gave similar values (Table 1). These recombination frequencies are >100-fold above the background levels determined by electroporation of individual pSV2neoX(B) or pneoAn plasmids (data not shown), as seen previously (TAGHIAN and NICKOLOFF 1998 ).

Evidence for mismatch-specific and nick-directed repair:
G418^r colonies were isolated for each of 12 crosses, and recombinant products were rescued. We found that >70% of rescued plasmids had the structure expected from a nonconservative exchange between pSV2 neoX(B) and pneoAn. Mismatch-repair direction was scored in rescued plasmids by ApaI digestion or by oligonucleotide hybridization, an example of which is shown in Figure 2. An average of 25 products with correct structures was analyzed per cross to determine repair direction for each type of mismatch. Segregation analysis (described below) also yielded information about repair direction. The combined data from these two approaches are shown in Table 2.

View larger version (51K): In this window In a new window Download PPT slide   Figure 2. Single-base marker detection by hybridizing mismatched/matched oligonucleotide probes. Eleven rescued recombination products from a T-C mismatch cross were run in duplicate on 0.8% agarose gels, blotted and hybridized with 15-base oligonucleotide probes that match either the G-C form (G) or the T-A form (T). Autoradiograph shows supercoiled and relaxed circular plasmids. See MATERIALS AND METHODS for hybridization details.

In a previous study, mismatches located asymmetrically between predicted nicks were repaired with bias consistent with repair directed from the proximal nick (DENG and NICKOLOFF 1994 ). Similarly, two studies with multiple mismatches suggested that repair direction for various mismatches was generally proportional to their distance from DNA nicks (MILLER et al. 1997 ; TAGHIAN and NICKOLOFF 1998 ). In the present study the single-base mismatches are predicted to be 130 bp and 112 bp from nicks on the top and bottom strands, respectively. Therefore, if all repair was nick-directed, a mismatched base on the top strand, located 54% of the distance from the nick on the top strand [130 bp = 0.54 (130 bp + 112 bp)], would on average be retained 54% of the time (or lost 46% of the time). We plotted the data from Table 2 relative to the value expected if repair was solely determined by nicks (Figure 3). For three crosses there were significant departures from the value predicted for nick-directed repair. One pair of crosses displayed strand-independent biased repair: those predicted to yield G-T and T-G mismatches (assuming SSA mediated by a 5' to 3' exonuclease) were significantly biased toward G-C (P = 0.016) and C-G (P = 0.035), respectively. For combined G-T/T-G data, P = 0.01, indicating a clear strand-independent bias toward loss of T. Different mismatches are predicted for different exonuclease polarities, so the crosses predicted to yield A-C and C-A with a 5' to 3' exonuclease would yield G-T and T-G, respectively, with a 3' to 5' exonuclease. However, of this pair, only C-A displayed a significant repair bias (P = 0.003); no bias was seen with A-C (P = 0.597). The bias in C-A repair may reflect a sequence context-dependent C-A binding activity (see DISCUSSION). Thus, strand-independent repair bias was observed only with that pair of crosses predicted to yield G-T mismatches with a 5' to 3' exonuclease. This bias is consistent with the well-known G-T G-C repair systems in bacteria (SOHAIL et al. 1990 ; LIEB 1991 ) and in mammalian cells (BROWN and JIRICNY 1988 ; WIEBAUER and JIRICNY 1989 , WIEBAUER and JIRICNY 1990 ). These data provide genetic evidence for a 5' to 3' exonuclease and also suggest that both mismatch-specific and nick-directed repair contribute to the observed product spectra.

View larger version (21K): In this window In a new window Download PPT slide   Figure 3. Percentage retention of the base on the top strand [listed first in each mismatch; donated by pSV2-neoX(B)]. The horizontal line at 54% retention represents the predicted value if all mismatch repair were nick-directed. Data are from Table 2. Values significantly different from the nick-directed model (P < 0.05, binomial confidence limits) are marked with asterisks. Data for the eight hetero-mismatches are shown on the right, with solid circles showing repair favoring either C or G and open circles showing repair favoring A or T. Homo-mismatches are indicated by shaded circles.

All single-base mismatches are repaired efficiently in wild-type cells:
A prior study indicated that markers located 349–523 bases from flanking nicks occur in hDNA 50% of the time (TAGHIAN and NICKOLOFF 1998 ). We therefore expected that the position 430 markers, located only 130 bases or less from flanking nicks, would occur in hDNA at even higher rates (see below). Segregation analysis indicates that all mismatches are repaired at high efficiency in wild-type CHO K1c cells (Table 2). The mean rate of segregation for all mispairs was 8.3%, with no significant differences among mispairs. These efficiencies are quite high compared to those seen with closed circular DNA (BROWN and JIRICNY 1988 ), as discussed below.

Although efficient repair is one explanation for the low segregation rates seen in CHO K1c cells, some recombinants may form without position 430 being included in hDNA and would be incorrectly scored as "repaired." To investigate this possibility, we performed two crosses in mismatch repair-deficient clone B cells predicted to produce T-T and G-T mismatches. In contrast to the <10% segregation in CHO K1c cells, ~75% segregation was observed in the mutant clone B cells (Figure 4). Interestingly, the nonsegregating T-T mismatch products were evenly distributed (three products corrected to A-T, two corrected to T-A), whereas seven of seven nonsegregating G-T mismatch products carried G-C pairs at position 430. The observation that G-T repair efficiency is reduced, and yet retains significant bias, provides insight into the competition between proteins involved in nick-directed and G-T-specific mismatch repair (see below).

View larger version (34K): In this window In a new window Download PPT slide   Figure 4. Percentage repair of G-T and T-T mismatches in wild-type K1c and mismatch repair-deficient clone B cells. Data for K1c cells are from Table 2. G-T and T-T data for clone B cells are from 26 and 24 products, respectively. Seven of seven G-T mismatches repaired in clone B cells were corrected to G-C.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Most extrachromosomal recombination in mammalian cells is thought to proceed by a nonconservative mechanism that yields apparent crossover products (SEIDMAN 1987 ; LIN et al. 1990 ). Such products in theory could arise by exchange in which hDNA forms infrequently and/or is limited in extent (DESAUTELS et al. 1990 ), or via intermediates with extensive hDNA formed by SSA as seen in three prior studies (DENG and NICKOLOFF 1994 ; MILLER et al. 1997 ; TAGHIAN and NICKOLOFF 1998 ). This study provides two lines of evidence for frequent hDNA. First, three crosses (C-A, G-T, and T-G; Figure 3) exhibited marker retention/loss rates that deviated from values expected if hDNA did not form. Second, segregation analysis in clone B cells indicated hDNA forms at least 75% of the time (Figure 4). Furthermore, although 25% of the nonsegregating products might have arisen in the absence of hDNA, one would expect that in this case the centrally located marker would be evenly distributed between the two parental configurations. Instead, seven of seven nonsegregating G-T products in clone B cells had G-C base pairs. The probability of obtaining this result in the absence of hDNA is very low (P = 0.5⁷ x 2 = 0.016), suggesting that most or all of these products arose by biased repair of hDNA (likely mediated by G-T glycosylase; see below) and that SSA produces hDNA at position 430 at rates approaching 100%.

In E. coli, G-A mispairs are repaired to G-C by the MutY adenine-specific DNA glycosylase (AU et al. 1989 ). The identification of a human mutY homologue (MYH) suggests that a similar mechanism may operate in mammalian cells (MCGOLDRICK et al. 1995 ). Biased G-A to G-C repair was observed in circular substrates (BROWN and JIRICNY 1988 ) and recombination intermediates (MILLER et al. 1997 ), but no G-A or A-G repair biases were found in this study, possibly reflecting sequence context effects (see below). The mammalian all-type endonuclease can nick all single-base mismatches with different efficiencies (YEH et al. 1991 ), but its function is unclear, and it may not be involved in mismatch correction (GALLINARI et al. 1998 ). Although we found most single-base mismatches were repaired without significant bias, there was a general trend toward retention of either C or G vs. A or T in the four pairs of hetero-mismatched bases (Figure 3), as seen with circular substrates in monkey cells (BROWN and JIRICNY 1988 ).

We observed differential repair biases of C-A and A-C mismatches. C-A was preferentially repaired to C-G (81%), and A-C G-C showed no bias (Figure 3). In our system, changing the orientation of hetero-mismatches changes the sequence context (Figure 1). C-A-specific repair may involve a protein analogous to a human protein that displays context-sensitive binding to C-A mismatches. The C-A binding protein recognized C-A mismatches in an ApG context, but not an ApA context (italics denote mismatched base; O'REGAN et al. 1996 ). Consistent with this, the C-A repair bias was seen in an ApG context, whereas no bias was seen for the A-C mismatch in an ApC context. Thus, the enzyme may only recognize C-A mismatches in the ApG context (ApT has not yet been tested). O'REGAN et al. 1996 suggested that sequence context-sensitive mismatch-specific repair systems form a complementary repair network. Sequence context effects may reflect different thermodynamic stabilities and helix melting enthalpies of mismatched bases (ABOUL-ELA et al. 1985 ; WERNTGES et al. 1986 ). Differential repair may also result from transcriptional asymmetry. In bacteria, repair efficiencies of mutY-dependent repair of G-A and C-A mismatches to G-C and C-G, respectively, were influenced by both sequence context and transcription (FOX et al. 1994 ).

The most prevalent mismatch is G-T, arising by deamination of 5-methylcytosine. Several studies in eukaryotic systems have demonstrated biased G-T G-C repair (BROWN and JIRICNY 1988 ; HEYWOOD and BURKE 1990 ; HOLMES et al. 1990 ) showing little strand bias (BROWN and JIRICNY 1987 ; HOLMES et al. 1990 ), although such a repair proclivity was not always seen (VARLET et al. 1990 ). G-T-specific thymine glycosylase is known to mediate G-T G-C repair (WIEBAUER and JIRICNY 1990 ; NEDDERMANN and JIRICNY 1993 ; GALLINARI et al. 1998 ). In vertebrates, 5-methylcytosine typically occurs in CpG dinucleotides. Human cell extracts and purified human G-T glycosylase recognize G-T mismatches most efficiently in a CpG context (GRIFFIN and KARRAN 1993 ; SIBGHAT-ULLAH and DAY 1995 ; SIBGHAT-ULLAH et al. 1996 ), consistent with a role for G-T glycosylase in counteracting G-T mismatches in this context. Similar context effects may modulate G-T processing in recombination intermediates in vivo. We found somewhat greater bias for G-T in a CpG context (75%) than for T-G in a CpC context (64%; Figure 1), although this difference was not statistically significant with these sample sizes. The purified G-T glycosylase also recognizes and removes thymine from other mismatches, including T-T, but at a lower efficiency (NEDDERMANN and JIRICNY 1993 , NEDDERMANN and JIRICNY 1994 ). The residual repair of T-T in clone B cells (Figure 4) may therefore be mediated by the G-T glycosylase as well.

With at least two systems capable of acting on G-T mismatches, the question arises whether the systems are in competition. GALLINARI et al. 1998 argued that G-T mismatches arising by deamination are not processed by a nick-directed pathway (i.e., postreplication mismatch repair) because it is unlikely that nicks will occur near sites of deamination. They further argued that G-T-specific repair does not interfere with nick-directed repair of G-T mismatches formed by replication errors. In contrast, our data suggest that nick-directed and G-T-specific repair do compete during processing of recombination intermediates. Our observed 64 and 75% biases toward G-C for the two G-T crosses are lower than the 85 and 93% biases seen with circular DNA (BROWN and JIRICNY 1987 ), consistent with the idea that nick-directed repair reduces bias imposed by G-T-specific repair. The completely biased (albeit low-efficiency) repair of G-T G-C in clone B cells lends further support to this idea. This low-efficiency repair would appear to be in conflict with the observation that these cells lack detectable G-T binding activity (BRANCH et al. 1993 ). However, clone B cells presumably lack Msh2 (AQUILINA et al. 1994 ), and most G-T binding may be Msh-dependent as these proteins were reported to be more abundant and bind more efficiently to G-T mismatches than G-T glycosylase (SIBGHAT-ULLAH et al. 1996 ). Previously, we concluded that biased repair of palindromic loop mismatches is also reduced by nick-directed repair. Interestingly, palindromic loop repair is highly efficient in both wild-type and clone B cells (TAGHIAN and NICKOLOFF 1998 ), whereas G-T repair in clone B cells is reduced to 27% of wild-type levels (Figure 4). Thus, G-T and loop mismatches are repaired by distinct mismatch-specific systems, both of which compete with nick-directed repair in recombination intermediates.

A key prediction of the SSA model is the presence of nicks (or single-stranded gaps) on opposite strands flanking hDNA in recombination intermediates. Previous studies of such intermediates with multiple mismatches indicated that repair was highly efficient (MILLER et al. 1997 ; TAGHIAN and NICKOLOFF 1998 ). However, these studies could not distinguish whether repair efficiencies of particular mismatches were influenced by co-repair at adjacent mismatches. We now report an overall repair efficiency of 92% for all types of individual single-base mismatches in recombination intermediates, with no significant differences among mismatch types (Table 2). These results contrast with prior studies showing differential repair of various single-base mismatches in eukaryotic cells or cell extracts (MUSTER-NASSAL and KOLODNER 1986 ; HOLMES et al. 1990 ; VARLET et al. 1990 ). In particular, A-G and T-T mismatches in circular substrates were repaired at a low rate (39%) in simian kidney cells (BROWN and JIRICNY 1988 ). Repair in HeLa cell extracts was more efficient with nicked substrates than with circular substrates (HOLMES et al. 1990 ; THOMAS et al. 1991 ). We did not directly compare repair between closed circular DNA and nicked DNA, but our data are nevertheless consistent with the idea that nicks enhance repair because our 8% average segregation rate is fourfold lower than the average for closed circular DNA (BROWN and JIRICNY 1988 ). Thus, studies with closed circular DNA, or singly nicked substrates, demonstrate that repair efficiency varies among mismatch types and that repair direction is influenced by which strand contains a nick, whereas our data indicate that flanking nicks in recombination intermediates direct efficient repair of all single-base mismatches, including those found to be poorly repaired in the absence of nicks, such as T-T and A-G.

Long-patch mismatch repair in bacteria involves (1) mismatch binding by MutS; (2) signaling from the bound mismatch to MutH (possibly involving MutL); (3) nicking at a hemi-methylated GATC site by MutH; and (4) excision repair directed from the nick. Although certain mismatches frequently escape repair in E. coli and yeast, such as C-C (BISHOP et al. 1989 ; KRAMER et al. 1989A ; LICHTEN et al. 1990 ; DETLOFF et al. 1991 ) and palindromic loop mismatches (NAG et al. 1989 ; NAG and PETES 1991 ; WENG and NICKOLOFF 1998 ), recent studies in yeast indicate that, at least for palindromic loop mismatches, poor repair is not due to inefficient mismatch recognition (MOORE et al. 1988 ; NAG et al. 1989 ; ALANI 1996 ; MANIVASAKAM et al. 1996 ). For example, although single-base and palindromic loop mismatches are both bound by yeast Msh2p-Msh6p (ALANI 1996 ), binding is modulated differently by adenosine triphosphate, providing a potential clue as to why these are differentially repaired. If mismatch repair in mammalian cells parallels that in yeast and E. coli, then the different repair efficiencies seen in nicked and non-nicked substrates in mammalian cells may not reflect differences in mismatch binding. Instead, the rate-limiting step may involve signaling to a nicking enzyme. Our data suggest that when nicks are provided, all mismatches are repaired efficiently.

	ACKNOWLEDGMENTS

We thank E. MILLER, D. TAGHIAN, and E. ALANI for helpful comments. This research was supported by grant CA-54079 to J. NICKOLOFF from the National Cancer Institute, National Institutes of Health.

Manuscript received February 18, 1998; Accepted for publication April 27, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ABOUL-ELA, F., D. KOH, and I. TINOCO, 1985  Base-base mismatches. Thermodynamics of double-helix formation for dCA₃XA₃G + dCT₃YT₃G (X, Y = A, C, G, T). Nucleic Acids Res. 13:4811-4824[Abstract/Free Full Text].

ALANI, E., 1996  The Saccharomyces cerevisiae Msh2 and Msh6 proteins form a complex that specifically binds to duplex oligonucleotides containing mismatched DNA base pairs. Mol. Cell. Biol. 16:5604-5615[Abstract].

AQUILINA, G., P. HESS, P. BRANCH, C. MACGEOCH, and I. CASCIANO et al., 1994  A mismatch recognition defect in colon carcinoma confers DNA microsatellite instability and a mutator phenotype. Proc. Natl. Acad. Sci. USA 91:8905-8909[Abstract/Free Full Text].

AU, K. G., S. CLARK, J. H. MILLER, and P. MODRICH, 1989  Escherichia coli mutY gene encodes an adenine glycosylase active on G/A mispairs. Proc. Natl. Acad. Sci. USA 86:8877-8881[Abstract/Free Full Text].

BISHOP, D. K., J. ANDERSEN, and R. D. KOLODNER, 1989  Specificity of mismatch repair following transformation of Saccharomyces cerevisiae with heteroduplex plasmid DNA. Proc. Natl. Acad. Sci. USA 86:3713-3717[Abstract/Free Full Text].

BOYER, J. C., A. UMAR, J. I. RISINGER, J. R. LIPFORD, and M. KANE et al., 1995  Microsatellite instability, mismatch repair deficiency, and genetic defects in human cancer cell lines. Cancer Res. 55:6063-6070[Abstract/Free Full Text].

BRANCH, P., G. AQUILINA, M. BIGNAMI, and P. KARRAN, 1993  Defective mismatch binding and a mutator phenotype in cells tolerant to DNA damage. Nature 362:652-654[Medline].

BRONNER, C. E., S. M. BAKER, P. T. MORRISON, G. WARREN, and L. G. SMITH et al., 1994  Mutation in the DNA mismatch repair gene homologue hMLH1 is associated with non-polyposis colon cancer. Nature 368:258-261[Medline].

BROOKMAN, K. W., J. E. LAMERDIN, M. P. THELEN, M. HWANG, and J. T. REARDON et al., 1996  ERCC4 (XPF) encodes a human nucleotide excision-repair protein with eukaryotic recombination homologs. Mol. Cell. Biol. 16:6553-6562[Abstract].

BROOKS, P., C. DOHET, G. ALMOUZNI, M. MECHALI, and M. RADMAN, 1989  Mismatch repair involving localized DNA synthesis in extracts of Xenopus eggs. Proc. Natl. Acad. Sci. USA 86:4425-4429[Abstract/Free Full Text].

BROWN, T. C. and J. JIRICNY, 1987  A specific mismatch repair event protects mammalian cells from loss of 5-methylcytosine. Cell 50:945-950[Medline].

BROWN, T. C. and J. JIRICNY, 1988  Different base/base mispairs are corrected with different efficiencies and specificities in monkey kidney cells. Cell 54:705-711[Medline].

CARRAWAY, M. and M. G. MARINUS, 1993  Repair of heteroduplex DNA molecules with multibase loops in Escherichia coli.. J. Bacteriol. 175:3972-3980[Abstract/Free Full Text].

CARROLL, D., C. W. LEHMAN, S. JEONG-YU, P. DOHRMANN, and R. J. DAWSON et al., 1994  Distribution of exchanges upon homologous recombination of exogenous DNA in Xenopus laevis oocytes. Genetics 138:445-457[Abstract].

CHURCH, G. M. and W. GILBERT, 1984  Genomic sequencing. Proc. Natl. Acad. Sci. USA 81:1991-1995[Abstract/Free Full Text].

DENG, W. P. and J. A. NICKOLOFF, 1994  Mismatch repair of heteroduplex DNA intermediates of extrachromosomal recombination in mammalian cells. Mol. Cell. Biol. 14:400-406[Abstract/Free Full Text].

DESAUTELS, L., S. BROUILLETTE, J. WALLENBURG, A. BELMAAZA, and N. GUSEW et al., 1990  Characterization of nonconservative homologous junctions in mammalian cells. Mol. Cell. Biol. 10:6613-6618[Abstract/Free Full Text].

DETLOFF, P., J. SIEBER, and T. D. PETES, 1991  Repair of specific base pair mismatches formed during meiotic recombination in the yeast Saccharomyces cerevisiae.. Mol. Cell. Biol. 11:737-745[Abstract/Free Full Text].

DE WIND, N., M. DEKKER, A. BERNS, M. RADMAN, and H. TE RIELE, 1995  Inactivation of the mouse Msh2 gene results in mismatch repair deficiency, methylation tolerance, hyperrecombination and predisposition to cancer. Cell 82:321-330[Medline].

DOEFLER, W., 1983  DNA methylation and gene activity. Annu. Rev. Biochem. 52:93-124[Medline].

FOLGER, K. R., K. THOMAS, and M. R. CAPECCHI, 1985  Efficient correction of mismatched bases in plasmid heteroduplexes injected into cultured mammalian cell nuclei. Mol. Cell. Biol. 5:70-74[Abstract/Free Full Text].

FOX, M. S., J. P. RADICELLA, and K. YAMAMOTO, 1994  Some features of base pair mismatch repair and its role in the formation of genetic recombinants. Experientia 50:253-260[Medline].

GALLINARI, P., P. NEDDERMANN and J. JIRICNY, 1998 Short patch mismatch repair in mammalian cells, pp. 119–131, in DNA Damage and Repair, Vol. II: DNA Repair in Higher Eukaryotes, edited by J. A. NICKOLOFF, and M. F. HOEKSTRA. Humana Press, Totowa, NJ.

GRIFFIN, S. and P. KARRAN, 1993  Incision of DNA at G-T mispairs by extracts of mammalian cells occurs preferentially at cytosine methylation sites and is not targeted by a separate G-T binding reaction. Biochemistry 32:13032-13039[Medline].

HARE, J. T. and J. H. TAYLOR, 1985  One role for DNA methylation in vertebrate cells is strand discrimination in mismatch repair. Proc. Natl. Acad. Sci. USA 82:7350-7354[Abstract/Free Full Text].

HENDERSON, G. and J. P. SIMONS, 1997  Processing of DNA prior to illegitimate recombination in mouse cells. Mol. Cell. Biol. 17:3779-3785[Abstract].

HESS, P., G. AQUILINA, E. DOGLIOTTI, and M. BIGNAMI, 1994  Spontaneous mutations at aprt locus in a mammalian cell line defective in mismatch recognition. Som. Cell Mol. Genet. 20:409-421[Medline].

HEYWOOD, L. A. and J. F. BURKE, 1990  Repair of single nucleotide DNA mismatches transfected into mammalian cells can occur by short-patch excision. Mutat. Res. 236:59-66[Medline].

HOLMES, J., JR., S. CLARK, and P. MODRICH, 1990  Strand-specific mismatch correction in nuclear extracts of human and Drosophila melanogaster cell lines. Proc. Natl. Acad. Sci. USA 87:5837-5841[Abstract/Free Full Text].

HUANG, K. N. and L. S. SYMINGTON, 1993  A 5'-3' exonuclease from Saccharomyces cerevisiae is required for in vitro recombination between linear DNA molecules with overlapping homology. Mol. Cell. Biol. 13:3125-3134[Abstract/Free Full Text].

IVANOV, E. L. and J. E. HABER, 1995  RAD1 and RAD10, but not other excision repair genes, are required for double-strand break-induced recombination in Saccharomyces cerevisiae.. Mol. Cell. Biol. 15:2245-2251[Abstract].

KAT, A., W. G. THILLY, W.-H. FANG, M. J. LONGLEY, and G.-M. LI et al., 1993  An alkylation-tolerant, mutator human cell line is deficient in strand-specific mismatch repair. Proc. Natl. Acad. Sci. USA 90:6424-6428[Abstract/Free Full Text].

KRAMER, B., W. KRAMER, M. S. WILLIAMSON, and S. FOGEL, 1989a  Heteroduplex DNA correction in Saccharomyces cerevisiae is mismatch specific and requires functional PMS genes. Mol. Cell. Biol. 9:4432-4440[Abstract/Free Full Text].

KRAMER, W., B. KRAMER, M. S. WILLIAMSON, and S. FOGEL, 1989b  Cloning and nucleotide sequence of DNA mismatch repair gene PMS1 from Saccharomyces cerevisiae: homology of PMS1 to procaryotic MutL and HexB. J. Bacteriol. 171:5339-5346[Abstract/Free Full Text].

LEVINSON, G. and G. A. GUTMAN, 1987  High frequencies of short frameshifts in poly-CA/TG tandem repeats borne by bacteriophage M13 in Escherichia coli.. Nucleic Acids Res. 15:5523-5538.

LICHTEN, M., C. GOYON, N. P. SCHULTES, D. TRECO, and J. W. SZOSTAK et al., 1990  Detection of heteroduplex DNA molecules among the products of Saccharomyces cerevisiae meiosis. Proc. Natl. Acad. Sci. USA 87:7653-7657[Abstract/Free Full Text].

LIEB, M., 1983  Specific mismatch correction in bacteriophage lambda crosses by very short patch repair. Mol. Gen. Genet. 191:118-125[Medline].

LIEB, M., 1991  Spontaneous mutation at a 5-methylcytosine hotspot is prevented by very short patch (VSP) mismatch repair. Genetics 128:23-27[Abstract].

LIN, F.-L., K. SPERLE, and N. STERNBERG, 1987  Extrachromosomal recombination in mammalian cells as studied with single- and double-stranded DNA substrates. Mol. Cell. Biol. 7:129-140[Abstract/Free Full Text].

LIN, F.-L., K. SPERLE, and N. STERNBERG, 1990  Intermolecular recombination between DNAs introduced into mouse L cells is mediated by a nonconservative pathway that leads to crossover products. Mol. Cell. Biol. 10:103-112[Abstract/Free Full Text].

MANIVASAKAM, P., S. M. ROSENBERG, and P. J. HASTINGS, 1996  Poorly repaired mismatches in heteroduplex DNA are hyper-recombinagenic in Saccharomyces cerevisiae.. Genetics 142:407-416[Abstract].

MCGOLDRICK, J. P., Y.-C. YEH, M. SOLOMON, J. M. ESSIGMANN, and A.-L. LU, 1995  Characterization of a mammalian homolog of the Escherichia coli MutY mismatch repair protein. Mol. Cell. Biol. 15:989-996[Abstract].

MILLER, E. M., H. L. HOUGH, J. W. CHO, and J. A. NICKOLOFF, 1997  Mismatch repair by efficient nick-directed, and less efficient mismatch-specific mechanisms in homologous recombination intermediates in Chinese hamster ovary cells. Genetics 147:743-753[Abstract].

MODRICH, P., 1991  Mechanisms and biological effects of mismatch repair. Annu. Rev. Genet. 25:229-253[Medline].

MOORE, C. W., D. M. HAMPSEY, J. F. ERNST, and F. SHERMAN, 1988  Differential mismatch repair can explain the disproportionalities between physical distances and recombination frequencies of cyc1 mutations in yeast. Genetics 119:21-34[Abstract/Free Full Text].

MUSTER-NASSAL, C. and R. KOLODNER, 1986  Mismatch correction catalyzed by cell-free extracts of Saccharomyces cerevisiae.. Proc. Natl. Acad. Sci. USA 83:7618-7622[Abstract/Free Full Text].

NAG, D. K. and T. D. PETES, 1991  Seven-base-pair inverted repeats in DNA form stable hairpins in vivo in Saccharomyces cerevisiae.. Genetics 129:669-673[Abstract].

NAG, D. K., M. A. WHITE, and T. D. PETES, 1989  Palindromic sequences in heteroduplex DNA inhibit mismatch repair in yeast. Nature 340:318-320[Medline].

NEDDERMANN, P. and J. JIRICNY, 1993  The purification of a mismatch-specific thymine-DNA glycosylase from HeLa cells. J. Biol. Chem. 268:21218-21224[Abstract/Free Full Text].

NEDDERMANN, P. and J. JIRICNY, 1994  Efficient removal of uracil from G-U mispairs by the mismatch-specific thymine DNA glycosylase from HeLa cells. Proc. Natl. Acad. Sci. USA 91:1642-1646[Abstract/Free Full Text].

NGHIEM, Y., M. CABRERA, C. G. CUPPLES, and J. H. MILLER, 1988  The MutY gene: a mutator locus in Escherichia coli that generates G:C to T:A transversions. Proc. Natl. Acad. Sci. USA 85:2709-2713[Abstract/Free Full Text].

NICKOLOFF, J. A., 1994  Sepharose spin column chromatography: a fast, nontoxic replacement for phenol:chloroform extraction/ethanol precipitation. Mol. Biotech. 1:105-108[Medline].

NICKOLOFF, J. A. and R. B. HALLICK, 1982  Synthetic deoxyoligonucleotides as general probes for chloroplast transfer RNA genes. Nucleic Acids Res. 10:8191-8210[Abstract/Free Full Text].

O'REGAN, N. E., P. BRANCH, P. MACPHERSON, and P. KARRAN, 1996  hMSH2-independent DNA mismatch recognition by human proteins. J. Biol. Chem. 271:1789-1796[Abstract/Free Full Text].

PETES, T. D., R. E. MALONE and L. S. SYMINGTON, 1991 Recombination in yeast, pp. 407–521 in The Molecular and Cellular Biology of the Yeast Saccharomyces: Genome Dynamics, Protein Synthesis, and Energetics, edited by J. R. BROACH, J. R. PRINGLE and E. W. JONES. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

PROLLA, T., S. BAKER and R. M. LISKAY, 1998 Genetics of mismatch repair, microsatellite instability and cancer, pp. 443–464 in DNA Damage and Repair, Vol. II: DNA Repair in Higher Eukaryotes, edited by J. A. NICKOLOFF and M. F. HOEKSTRA. Humana Press, Totowa, NJ.

RASMUSSEN, L. J., L. SAMSON and M. G. MARINUS, 1998 Dam-directed DNA mismatch repair, pp. 205–228 in DNA Damage and Repair, Vol. I: DNA Repair in Prokaryotes and Lower Eukaryotes, edited by J. A. NICKOLOFF and M. F. HOEKSTRA. Humana Press, Totowa, NJ.

REENAN, R. A. G. and R. D. KOLODNER, 1992  Isolation and characterization of two Saccharomyces cerevisiae genes encoding homologs of the bacterial HexA and MutS mismatch repair proteins. Genetics 132:963-973[Abstract].

RIGGS, A. D. and P. A. JONES, 1983  5-Methylcytosine regulation and cancer. Adv. Cancer Res. 40:1-30[Medline].

SAMBROOK, J., E. F. FRITSCH and T. MANIATIS, 1989 Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SCHMUTTE, C., A. S. YANG, R. W. BEART, and P. A. JONES, 1995  Base excision repair of U:G mismatches at a mutational hotspot in the p53 gene is more efficient than base excision repair of T:G mismatches in extracts of human colon tumors. Cancer Res. 55:3742-3746[Abstract/Free Full Text].

SEIDMAN, M. M., 1987  Intermolecular homologous recombination between transfected sequences in mammalian cells is primarily nonconservative. Mol. Cell. Biol. 7:3561-3565[Abstract/Free Full Text].

SHEN, J.-C., W. M. RIDEOUT, and P. A. JONES, 1994  The rate of hydrolytic deamination of 5-methylcytosine in double-stranded DNA. Nucleic Acids Res. 22:972-976[Abstract/Free Full Text].

SIBGHAT-ULLAH, and R. S. DAY, 1995  Site-specificity of incisions at G-T and O-6-methylguanine-T base mismatches in DNA by human cell-free extracts. Biochemistry 34:6869-6875[Medline].

SIBGHAT-ULLAH, Y.-Z., P. GALLINARI, Y.-Z. XU, M. F. GOODMAN, and L. B. BLOOM et al., 1996  Base analog and neighboring base effects on substrate specificity of recombinant human G:T mismatch-specific thymine DNA-glycosylase. Biochemistry 35:12926-12932[Medline].

SOHAIL, A., M. LIEB, M. DAR, and A. S. BHAGWAT, 1990  A gene required for very short patch repair in Escherchia coli is adjacent to the DNA cytosine methylase gene. J. Bacteriol. 172:4214-4221[Abstract/Free Full Text].

SOUTHERN, P. J. and P. BERG, 1982  Transformation of mammalian cells to antibiotic resistance with a bacterial gene under control of the SV40 early region promoter. J. Mol. Appl. Genet. 1:327-341[Medline].

STRAND, M., T. A. PROLLA, R. M. LISKAY, and T. D. PETES, 1993  Destabilization of tracts of simple repetitive DNA in yeast by mutations affecting DNA mismatch repair. Nature 365:207-208[Medline].

SUN, H., D. TRECO, and J. W. SZOSTAK, 1991  Extensive 3'-overhanging, single-stranded DNA associated with meiosis-specific double-strand breaks at the ARG4 recombination initiation site. Cell 64:1155-1161[Medline].

TAGHIAN, D. G., and J. A. NICKOLOFF, 1995 Electrotransformation of Chinese hamster ovary cells, pp. 115–121 in Animal Cell Electroporation and Electrofusion Protocols, edited by J. A. NICKOLOFF. Humana Press, Totowa, NJ.

TAGHIAN, D. G. and J. A. NICKOLOFF, 1998  Biased short tract repair of palindromic loop mismatches in mammalian cells. Genetics 148:1257-1268[Abstract/Free Full Text].

THOMAS, D. C., J. D. ROBERTS, and T. A. KUNKEL, 1991  Heteroduplex repair in extracts of human HeLa cells. J. Biol. Chem. 266:3744-3751[Abstract/Free Full Text].

TSAI-WU, J. J., J. P. RADICELLA, and A.-L. LU, 1991  Nucleotide sequence of the Escherichia coli micA gene required for A/G-specific mismatch repair: identity of MicA and MutY. J. Bacteriol. 173:1902-1910[Abstract/Free Full Text].

UMAR, A., J. C. BOYER, and T. A. KUNKEL, 1994  DNA loop repair by human cell extracts. Science 266:814-816[Abstract/Free Full Text].

VARLET, I., M. RADMAN, and P. BROOKS, 1990  DNA mismatch repair in Xenopus egg extracts: repair efficiency and DNA repair synthesis for all single base-pair mismatches. Proc. Natl. Acad. Sci. USA 87:7883-7887[Abstract/Free Full Text].

WANG, Z. and T. G. ROSSMAN, 1994  Large-scale supercoiled plasmid preparation by acidic phenol extraction. Biotechniques 16:460-463[Medline]