Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Jan 11, 2011; 108(2): 698–703.
Published online Dec 21, 2010. doi:  10.1073/pnas.1012363108
PMCID: PMC3021083

Role for topoisomerase 1 in transcription-associated mutagenesis in yeast


High levels of transcription in Saccharomyces cerevisiae are associated with increased genetic instability, which has been linked to DNA damage. Here, we describe a pGAL-CAN1 forward mutation assay for studying transcription-associated mutagenesis (TAM) in yeast. In a wild-type background with no alterations in DNA repair capacity, ≈50% of forward mutations that arise in the CAN1 gene under high-transcription conditions are deletions of 2–5 bp. Furthermore, the deletions characteristic of TAM localize to discrete hotspots that coincide with 2–4 copies of a tandem repeat. Although the signature deletions of TAM are not affected by the loss of error-free or error-prone lesion bypass pathways, they are completely eliminated by deletion of the TOP1 gene, which encodes the yeast type IB topoisomerase. Hotspots can be transposed into the context of a frameshift reversion assay, which is sensitive enough to detect Top1-dependent deletions even in the absence of high transcription. We suggest that the accumulation of Top1 cleavage complexes is related to the level of transcription and that their removal leads to the signature deletions. Given the high degree of conservation between DNA metabolic processes, the links established here among transcription, Top1, and mutagenesis are likely to extend beyond the yeast system.

The basic DNA metabolic processes of transcription, replication, repair, and recombination are usually considered in isolation, yet they have the potential to influence one another in terms of frequency, efficiency, and fidelity. In the yeast Saccharomyces cerevisiae, high levels of transcription have been shown to have strong stimulatory effects on homologous recombination and mutagenesis and, hence, are associated with genome instability (reviewed in refs. 1 and 2). Most analyses of transcription-associated mutagenesis and recombination (TAM and TAR, respectively) in yeast have focused on identifying genes whose loss enhances or suppresses these processes. A general theme that has emerged from such studies is that transcription leads to diverse types of DNA damage, which can be potent initiators for recombination and/or mutagenesis. It has been suggested, for example, that transcription increases the single-stranded character of the nontranscribed strand, which may promote chemical reactivity or susceptibility to endogenous DNA-damaging agents (3, 4). Such enhanced single-strandedness could reflect exclusion of the nontranscribed strand from the transient RNA–DNA hybrids within transcription bubbles that track with RNA polymerase, or the more stable RNA–DNA hybrids (R-loops) that can form in the wake of RNA polymerase. R-loops become particularly problematic when mRNA maturation is perturbed; their persistence is associated with replication fork collapse, and they can thus be an important source of TAR (5). In experiments designed to pinpoint specific types of DNA damage that lead to TAM, we unexpectedly found that high levels of transcription result in a localized replacement of thymine with the RNA-specific base uracil, the removal of which generates mutagenic abasic sites (6). The replacement of thymine with uracil suggests that there may be a more general, transcription-associated perturbation of the nucleotide pools that support DNA synthesis, which also could drive mutagenesis by elevating levels of nucleotide misincorporation. Finally, a failure to remove the helical stress associated with transcription can destabilize repetitive DNA sequences, presumably by elevating the level of recombination-initiating nicks and breaks (7).

Studies of TAM in yeast have generally used frameshift reversion assays and have focused on how rates and spectra are altered in various repair-defective backgrounds (6, 8, 9). Although such assays have been tremendously useful, there are stringent limitations on the mutation types that can be detected and on their location relative to the original frameshift mutation. To obtain a more unbiased assessment of DNA instability under high-transcription conditions, we examined the effect of transcription on forward mutations in the LYS2 gene (10). Although transcription appeared to elevate most mutation types in this system, 2-bp deletions emerged as a clear signature of TAM. In this study, we further explore the source and the nature of the unique 2-bp deletion class. Using a reporter based on CAN1, we find that the 2-bp deletions characteristic of TAM accumulate at discrete hotspots and that they absolutely require the presence of topoisomerase 1 (Top1), an enzyme that specifically removes transcription-associated supercoils (11). We suggest that irreversible Top1 cleavage complexes accumulate in highly transcribed DNA and that their removal generates the distinctive 2-bp deletion signature. Given the high conservation of DNA metabolic processes, the transcription-associated link between topoisomerase activity and mutagenesis established here is expected to extend to higher eukaryotes. In particular, it may be relevant to the emergence of secondary tumors following chemotherapy with Top1 inhibitors (12).


The large size of the LYS2 gene used to identify signature TAM events limits the use of this assay for discerning and studying mutation hotspots. In this study, a high-transcription version of the CAN1 gene was constructed by replacing the endogenous CAN1 promoter (pCAN) with a highly inducible, galactose-regulated promoter (pGAL). The replacement was done in a background lacking the Gal80 repressor of pGAL, resulting in constitutive, high-level expression from this promoter. Hereafter, the pCAN-CAN1 gal80 and pGAL-CAN1 gal80 strains are referred to as low- and high-transcription strains, respectively. The CAN1 locus encodes arginine permease, which in addition to arginine, also transports the toxic analog canavanine. Wild-type cells are thus sensitive to canavanine, and any forward mutation that disrupts function of the encoded protein results in a canavanine-resistant (Can-R) phenotype.

CAN1 Forward Mutation Rates and Spectra Under Low- and High-Transcription Conditions.

In a repair-proficient background, high levels of transcription through the CAN1 gene resulted in a 12.3-fold increase in the rate of Can-R colonies (Table 1). To determine the types of mutations stimulated by high levels of transcription, we sequenced the CAN1 ORF in independent mutants isolated from the low- and high-transcription strains (Fig. S1 and Fig. S2, respectively, and Table S1). Base substitutions and 1-bp deletions were the most abundant mutation classes under low-transcription conditions, comprising 83% (68/82) and 9% (7/82) of the spectrum, respectively (Table 1). Although there was a similar proportion of 1-bp deletions (14/91 = 15%) under high-transcription conditions, base substitutions accounted for only 19% (17/91) of the spectrum. The high-transcription spectrum was instead dominated by short deletions of 2–5 bp, which comprised an astounding 55% (50/91) of the events recovered. Only two such events were observed under low-transcription conditions, corresponding to an ≈300-fold increase in 2- to 5-bp deletions when CAN1 is highly transcribed.

Table 1.
CAN1 forward mutations recovered under low- and high-transcription conditions

Rates of Small Deletions Are Not Altered in rad51Δ or rev3Δ Strains.

Previous analyses of TAM in excision repair-defective backgrounds suggest that most mutagenesis is due to the accumulation of transcription-associated DNA damage (6, 8, 9). Consistent with this interpretation, loss of homologous recombination (HR), an error-free pathway of damage bypass, elevates TAM in frameshift reversion assays (8, 13). Loss of translesion synthesis (TLS), which is an alternative, error-prone pathway, has the reverse effect and lowers TAM in these assays (6, 9, 13). We thus examined the effect of eliminating the Rad51 protein, which is required for the strand invasion step of HR (14), or the Rev3 protein, which is the catalytic subunit of the Polζ TLS polymerase (15), on TAM in the pGAL-CAN1 system. As expected, loss of HR or TLS resulted in an increase or decrease, respectively, in the overall rate of TAM (Table 1). Although base substitution and −1 frameshift rates changed in concert with the overall mutation rate, the rate of the 2- to 5-bp deletion class was not altered upon loss of either HR or TLS. These results indicate that short deletions are derived from a unique type of transcription-associated damage that cannot be readily bypassed by canonical error-free or error-prone pathways.

Short Deletions Decrease in top1Δ Cells Under High-Transcription Conditions.

We proposed that the transcription-dependent, 2-bp deletions detected in the pGAL-LYS2 assay might reflect the processing of topoisomerase-generated nicks or breaks (10). To test this hypothesis, we examined the rates and spectra of CAN1 forward mutations in low- and high-transcription top1Δ strains (Table 1). Although loss of Top1 did not affect the rate of Can-R colonies under low-transcription conditions, there was a small but significant (twofold) reduction in the Can-R rate under high-transcription conditions. This reduction can be completely accounted for by the loss of a single class of transcription-associated mutations: 2- to 5-bp deletions. We estimate that the rate of these deletions was reduced approximately two orders of magnitude upon loss of Top1, whereas other mutation classes were not affected.

Given the key role that Top1 plays in relieving transcription-associated supercoiling, a trivial explanation for the disappearance of the 2- to 5-bp deletions in the top1Δ strain would be a decreased level of CAN1 transcription. The persistence of the other transcription-associated mutation classes in the top1Δ strain suggested that this explanation is not the case. Quantitative RT-PCR confirmed that CAN1 steady-state transcript levels were not affected by the presence/absence of Top1 under high-transcription conditions (Table S2).

Top1-Dependent Deletions Occur at Distinct Hotspots.

Fig. 1 presents the positions of 141 deletions of 2–5 bp identified in wild-type, rad51Δ, and rev3Δ strains under high-transcription conditions. The most abundant deletion class was 2-bp deletions, and 88% (89/101) of these removed a single unit of a 2- to 4-unit dinucleotide repeat. Strikingly, 2-bp deletions localized to only a subset of the 63 dinucleotide repeats found within the CAN1 ORF (highlighted in gray in Fig. 1). The strongest hotspot was an (AT)2 repeat at position 1127, where 25% (25/101) of the 2-bp deletions were found. Additional 2-bp deletion hotspots corresponded to a (TC)3 repeat beginning at position 1448, a (CT)3 repeat at position 1002, an (AT)2 repeat at position 275, and an (AT)2 repeat at position 375 (14, 10, 9, and 8 events, respectively). Although all six possible dinucleotide repeats are present within the CAN1 ORF, the major hotspots coincided with either an AT/TA or TC/CT repeat. There was thus neither a single dinucleotide sequence uniquely relevant to deletion formation nor were all dinucleotide sequences equivalent hotspots.

Fig. 1.
Short (2- to 5-bp) deletions in the pGAL-CAN1 spectrum. The entire CAN1 ORF is shown, and deletions are indicated above the relevant sequence as bars. The size of individual bars reflects the size of the corresponding deletion. Deletions isolated from ...

In addition to the 2-bp deletion hotspots, there was a hotspot for 3-bp deletions (15 events) at a (GTT)2 trinucleotide repeat beginning at position 970. Although this particular in-frame deletion appears to have deleted an essential valine residue, there may have been additional 3-bp deletion hotspots that failed to disrupt protein function and, hence, went undetected. Finally, there was a weak hotspot for 4-bp deletions (5 events) at position 1268, and this cluster coincided with an interrupted CAT trinucleotide repeat (CATACAT). Among all of the deletion hotspots detected, there were none that were overrepresented in the WT versus rev3Δ background. Too few small deletions were isolated from the rad51Δ background to allow a similar comparison.

To identify potential sequence similarities in addition to the observed association with short (primarily dinucleotide) repeats, all sites within CAN1 where at least two deletions were identified were ranked based on the number of events. These sites were then aligned by the beginning of the relevant repeat (Table S3). Although no consensus sequence was evident, we note that four of the six strongest hotspots overlap with a 5′-CAT-3′ motif and one contains a 5′-GTT-3′ motif, both of which match the weak consensus site for eukaryotic Top1 cleavage in vitro [5′-(not G)(G/C)(T/A)T-3′; ref. 16]. The overlap with potential Top1 cleavage sites is consistent with the genetic dependence of the small deletions on the presence of Top1. As will be elaborated in the Discussion, hotspots likely reflect the mechanism of deletion formation, which we suggest requires the colocalization of a Top1 cleavage site with a short repeat.

Top1-Dependent Deletion Hotspots Are Portable.

In a previous study, the relationship between transcript level and TAM was examined by using a frameshift allele under control of a tetracycline/doxycycline (Dox)-regulated promoter (pTET-lys2ΔA746 allele; ref. 9). Because reversion of this allele selects for net +1 frameshift mutations, we reasoned that it should efficiently detect the weakly clustered, 2-bp deletions seen in the pGAL-LYS2 forward mutation assay (10). This result did not occur, however, and the current analyses with the pGAL-CAN1 allele suggest an explanation. Namely, no hotspot for Top1-dependent deletions naturally exists within the relatively limited, 150-bp reversion window of the pTET-lys2ΔA746 allele. To verify this explanation and to examine whether hotspot sequences are portable, we inserted two of the Top1-dependent hotspots identified in the pGAL-CAN1 system into the reversion window of the lys2ΔA746NR (“no run”) allele, a modified version that contains no mononucleotide runs >3N (17). The lys2ΔA746NR,(AT)2 and lys2ΔA746NR,(TC)3 alleles thus constructed have ≈30 bp of additional sequence centered on the (AT)2 and (TC)3 hotspots beginning at positions 1127 and 1448 of the CAN1 ORF, respectively.

In the absence of a transposed hotspot, the Lys+ reversion rate increased 21-fold when the lys2ΔA746NR allele was highly transcribed (Fig. 2A and Table S4). Although 2-bp deletions comprised 20% of the high-transcription spectrum, the 17 events observed were spread over 13 sites within the reversion window (Fig. S3B), and their rate did not change upon deletion of the TOP1 gene. Surprisingly, addition of the (AT)2 or (TC)3 hotspot to the reversion window elevated the Lys+ rate eight- or threefold, respectively, even under low-transcription conditions (Fig. 2 B and C and Table S4). Consistent with the rate increases, the majority of events were 2-bp deletions within the dinucleotide repeat of the introduced hotspot (Fig. S3A). Combining high-transcription with the presence of either the (AT)2 or (TC)3 hotspot had a very strong, synergistic effect on the overall reversion rate, with the rate increasing 490- or 160-fold, respectively, relative to the reversion rate of the unmodified lys2ΔA746NR allele under low-transcription conditions (Fig. 2). Finally, 95% (88/93) and 88% (83/94) of events in the (AT)2 and (TC)3 high-transcription spectra, respectively, were 2-bp deletions at the transposed hotspot (Fig. S3B), whereas no hotspot events were recovered from the top1Δ high-transcription strains (Fig. 2 B and C and Fig. S3C). The high transcription rate increases conferred by the introduced hotspots thus completely depended on the presence of Top1. Apart from hotspot events, there were no other mutation types that were significantly affected by the presence/absence of Top1. As observed with the pGAL-CAN1 allele, the steady-state level of pTET-LYS2 transcript was indistinguishable in the TOP1 and top1Δ high-transcription strains (Table S2).

Fig. 2.
Reversion rates of lys2ΔA746NR alleles under high- and low-transcription conditions. LYS2 and TET promoters were used for low- and high-transcription conditions, respectively. White bars correspond to 2-bp deletions in the native LYS2 sequence, ...

Catalytic Activity of Top1 Is Required for Hotspot-Associated Deletions.

The known role of Top1 in relaxing transcription-generated supercoils implicates its strand-nicking activity in TAM. It is formally possible, however, that the signature deletions require only the physical presence of Top1, perhaps as a component of a larger complex. To directly test the relevance of Top1 catalytic activity to TAM, we examined the effect of changing the active-site tyrosine to phenylalanine on deletion accumulation at the transposed (AT)2 and (TC)3 hotspots. For this analysis, a plasmid expressing either the wild-type TOP1 gene or a catalytically dead top1-Y727F allele (18) was introduced into the appropriate top1Δ strains. Whereas the TOP1 allele fully restored TAM at both transposed hotspots, introduction of the top1-Y727F allele failed to increase mutagenesis at either (Table S5).

Tdp1 Does Not Affect the Rate of 2-Bp Deletions at Transposed Hotspots.

Yeast tyrosyl DNA phosphodiesterase (Tdp1) was identified biochemically, and its loss sensitizes cells to the toxic effects of camptothecin, a Top1-specific poison (19). Because we speculate that the Top1-dependent deletion hotspots reflect the repair of trapped Top1 cleavage complexes, it seemed likely that Tdp1 loss would be involved in either limiting or promoting these events. We thus deleted the TDP1 gene from strains containing the lys2ΔA746NR,(AT)2 or lys2ΔA746NR,(TC)3 allele and examined Lys+ rates and spectra under high transcription conditions. As shown in Fig. 2, both were indistinguishable in the WT and tdp1 backgrounds (see also Fig. S3 B and D).


We reported that high levels of transcription through a pGAL-LYS2 reporter generate a variety of forward mutations, with short deletions in tandem repeats being a novel signature of TAM (10). Using a unique pGAL-CAN1 assay, we demonstrate here that 2- to 5-bp deletions occur at distinct hotspots that coincide with short repeats. Although these specific events are not affected by the loss of DNA damage bypass pathways, they absolutely require the catalytic activity of Top1, a topoisomerase specifically implicated in removing transcription-associated supercoils (11). Finally, we demonstrate that representative Top1-dependent, 2-bp deletion hotspots are fully functional when transposed into a frameshift reversion assay, an assay sensitive enough to detect the signature 2-bp deletions even in the absence of high transcription. Their occurrence is greatly stimulated, however, by high transcription. Below, we discuss the implications of these results and propose a model for how removal of a trapped Top1 cleavage complex can generate the signature deletions of TAM.

The 2-bp deletions in the pGAL-CAN1 reporter were associated with only a subset of the short dinucleotide repeats present in the ORF (Fig. 1). Although a dinucleotide repeat appears to be necessary for the formation of 2-bp deletions, it is not sufficient. There is an absolute requirement for Top1, however, and many hotspots overlap with or are immediately adjacent to an identifiable Top1 consensus site. Although it is possible that Top1 nicks DNA preferentially at consensus sites that colocalize with a dinucleotide repeat, we think it is more likely that the repeat is important for stabilizing a Top1-dependent mutagenic intermediate. We note that there are numerous mononucleotide runs in the CAN1 ORF that should be able to, in principle, support similar deletion formation. The likely reason that none of these runs were hotspots for 2-bp deletions is that they simply do not colocalize with a strong Top1 cleavage site. Although this explanation also could account for the absence of Top1-dependent 1-bp deletions, it is not clear whether the mutagenic processing of a cleavage complex would generate the requisite 1-nt gap (see model below).

Top1 is a type IB topoisomerase that relaxes either positive or negative supercoils by nicking one strand of DNA, generating a covalent 3′-phosphotyrosyl link to the enzyme on one side of the nick and a free 5′-OH on the other side (reviewed in refs. 12 and 20). By a mechanism of constrained rotation, multiple supercoils can be removed before the enzyme reseals the nick. Top1 is a common target of chemotherapeutics such as camptothecin (CPT), which reversibly stabilizes the covalent Top1 intermediate. Toxicity of CPT requires DNA replication (21) and is greatly enhanced by the loss of HR (22), suggesting that cell death is triggered by double-strand breaks (DSBs) created when a replication fork encounters a stabilized Top1 cleavage complex. Alternatively, a replication-independent DSB could be generated if there is a nick on the strand complementary to the Top1-linked strand, or if there is a Top1 cleavage complex on the complementary strand. Short deletions could then be generated by microhomology-mediated nonhomologous end-joining (NHEJ; reviewed in ref. 23), thereby explaining their occurrence at hotspots that colocalize with tandem repeats. We do not favor such a model for two reasons. First, the signature TAM events in the pGAL-CAN1 allele were not elevated upon loss of HR (rad51Δ background), which is the predominant pathway for DSB repair in yeast. Second, the deletion events at the transposed hotspots were not affected by deletion of the DNL4 gene (Table S4), which encodes the ligase required for NHEJ. At least in our TAM assays, the mutagenic effects of Top1 appear to result from an intermediate distinct from the DSBs responsible for the toxic effects.

Fig. 3 presents a model for how removal of Top1 trapped at a nick can generate short deletions in the context of a tandem repeat. The specific sequence shown is that of the strongest (AT)2 hotspot for 2-bp deletions, which has an adjacent Top1 consensus site. In this model, the nick is expanded to a 2-nt gap that lies within the dinucleotide repeat by enzymes that generate ligatable 3′-OH and 5′-PO4 ends (with larger repeat units, the size of the gap would correspond to that of the repeat unit). These ends would then be brought together for ligation by misalignment between the complementary strands, generating a 2-nt loop on the intact strand. The resulting distortion would then either be removed by the mismatch repair machinery or resolved at the next round of replication. We note that Top1 itself can mediate ligation across gaps with a 5′-OH in vitro (24), but it is not clear how a comparable gap might be generated in vivo.

Fig. 3.
Model for Top1-dependent deletions. Sequence surrounding the (AT)2 hotspot is shown. A possible Top1 consensus sequence is highlighted in gray, and the ends generated by Top1 cleavage are as indicated. See Discussion for additional details.

In any model of Top1-dependent deletion formation, the enzyme must be removed from the 3′ end before ligation can occur. Multiple candidate proteins have been identified from genetic studies in yeast, all of which have focused on the regulation of CPT toxicity. Although we do not yet know which proteins are relevant to deletion formation at a nick, Tdp1 appears to neither enhance nor repress these events in our TAM assays. We note that this result is consistent with biochemical studies of Tdp1, where a phosphotyrosine at the end of a DSB is a preferred substrate relative to one at a nick (25). Tdp1 is the only known yeast protein that “cleanly” removes a Top1-derived peptide [the resulting 3′-PO4 can be redundantly converted to a 3′-OH by Tpp1, Apn1, or Apn2 (26)], with other candidates likely working as structure-specific endonucleases that concurrently remove adjacent sequence (2729). In the accompanying paper Takahashi et al. (30) examine the roles of the Mus81 and Rad1 endonucleases in Top1-dependent TAM. Their data suggest that these proteins play a role in generating the signature deletions, but neither is absolutely required. What processes the 5′-OH created by Top1 to a ligatable 5′-PO4 in yeast is not known (31). It could be directly phosphorylated by a kinase, removed by a 5′ > 3′ exonuclease, or removed as part of a 5′ flap generated by extension of the 3′-OH. Further genetic studies of proteins required for deletion formation and biochemical analysis of precisely where Top1 cleaves DNA relative to the genetically defined deletion hotspots will provide insight into whether nucleotides are likely deleted from one or both sides of the break before ligation.

The involvement of Top1 in TAM likely reflects the active recruitment of the enzyme to remove transcription-associated supercoils. Also relevant to deletion formation is the trapping of Top1, the efficiency of which may be concurrently enhanced under high-transcription conditions (30). An increased proportion of Top1 intermediates could be trapped by transcription-associated DNA damage and/or by collision with RNA polymerase (reviewed in ref. 12). Overexpression of Top1 sensitizes yeast to DNA damage (32), and a wide variety of nearby lesions have been shown to inhibit ligation following Top1-dependent cleavage of DNA in vitro. At least some of these lesions (e.g., abasic sites; ref. 33) are known to accumulate in highly transcribed DNA when repair pathways are compromised (6), but whether they might also accumulate to sufficient levels in a repair-competent background is not known. With regard to the process of transcription, CPT-stabilized Top1 cleavage complexes located specifically on the transcribed strand are converted to irreversible strand breaks by elongating RNA polymerase in vitro (34). It has been suggested that collision with RNA polymerase dislodges the free 5′-OH from the Top1 active site, thereby preventing (or reducing) subsequent ligation.

The demonstration that Top1 is responsible for the majority of TAM events in repair-competent yeast cells reveals an unexpected source of genome instability. Strikingly, the sizes of the deletions associated with Top1 activity reflect the sizes of the short tandem repeats where they cluster. The demonstration that discrete 2-bp deletion hotspots function in the context of a very sensitive reversion assay allows representative hotspots to be studied in isolation and will greatly facilitate future mechanistic study. Given the universal involvement of topoisomerases in relieving the superhelical stress associated with transcription, the results reported here are likely to extrapolate to higher eukaryotes. Whether these yeast results also will be relevant to the mutagenic effects of CPT-trapped Top1 in mammalian cell lines (35, 36) remains to be explored.

Materials and Methods

Strain Constructions.

SJR282 (MATα ade2-101oc his3Δ200 ura3ΔNco suc2 gal80Δ::HIS3; ref. 13) and YPH45 (MATa ura3-52 ade2-101oc trp1Δ1; ref. 37) derivatives were used for CAN1 forward mutation and lys2 reversion assays, respectively. The high-transcription pGAL-CAN1 strain was generated by replacing ≈50 bp immediately upstream of the CAN1 ORF with a PCR fragment containing the GAL1 promoter (pGAL). The PCR fragment was amplified by using pFA6a-kanMX6-PGAL1 (38) as a template and primers with 5′-terminal homology to CAN1 (forward primer 5′-GTTTTTAATCTGTCGTCAATCGAAAGTTTATTTCAGAGTTgaattcgagctcgtttaaac and reverse primer 5′-TATGCTTCTCCTCTATGTCGGCGTCTTCTTTTGAATTTGTCATtttgagatccgggtttt; CAN1 sequences are in uppercase). Although CAN1 overexpression can significantly impair growth in some strain backgrounds, little or no effect was observed in the SJR282 strain background used here.

Strains containing a constitutive his4Δ::LYS2 (SJR2259) or doxycycline-regulated his4Δ::pTET-LYS2 allele (SJR2261) in the same orientation as HIS4 were described (9). lys2ΔA746NR, lys2ΔA746NR,(AT)2, or lys2ΔA746NR,(TC)3 derivatives were constructed by two-step allele replacement by using AflII-digested pSR963 (17), pSR1002, or pSR1003, respectively. pSR1002 and pSR1003 were derived by annealing oligonucleotides [5′-gatcGGTTTTGCCACATATCTTCAACGCTG and 5′-gatcCAGCGTTGAAGATATGTGGCAAAACC for the (AT)2 hotspot or 5′-gatcATACCGTGGCATCTCTCGTGACGAGTTAC and 5′-gatcGTAACTCGTCACGAGAGATGCCACGGTAT for the (TC)3 hotspot; hotspots are underlined, and BglII-compatible overhangs are in lowercase] and then inserting the resulting ≈30 bp fragments into BglII-digested pSR963. RAD51, REV3, TOP1, TDP1, or DNL4 were deleted by using PCR-generated deletion cassettes amplified from plasmids containing appropriate drug-resistance markers. Table S6 gives a complete list of strains.

Mutation Rates and Spectra.

Cells were grown nonselectively in YEP (1% yeast extract, 2% Bacto-peptone, 250 μg/mL adenine hemisulfate; 2% agar for plates) medium supplemented with 2% dextrose (YEPD) or 2% glycerol and 2% ethanol (YEPGE). Selective growth was on synthetic complete dextrose medium lacking the appropriate nutrient (e.g., SCD-Lys). Canavanine-resistant colonies were selected on SCD-Arg plates containing 60 μg/mL l-canavanine sulfate. To determine mutation rates, cells from saturated YEPGE cultures were washed with H2O and appropriate dilutions plated on YEPD or selective medium to determine total or mutant cell numbers, respectively, in each culture. Data from at least 17 cultures were used to calculate each mutation rate by using the method of the median (39); 95% confidence intervals were determined as described (40).

Genomic DNAs from independent mutants were isolated by using a 96-well format (http://jinks-robertsonlab.duhs.duke.edu/protocols/yeast_prep.html), and the region of interest was sequenced either by the University of Washington High-Throughput Genomics Unit or by the Duke University DNA Analysis Facility. Rates of individual mutation types were determined by multiplying the total mutation rate by the proportion of the mutation type in the corresponding spectrum. A minority of can1 mutants (< 25% of the total sequenced for any given strain) did not contain a mutation within the ORF. Given their negligible effect, these were not taken into account when calculating the proportion of specific mutation types.

Supplementary Material

Supporting Information:


We acknowledge Mary-Ann Bjornsti (University of Alabama School of Medicine, Birmingham, AL) for providing plasmids containing the TOP1 or top1-Y727F allele and Serge Boiteux for communicating results prior to publication. Work in the laboratories of S.J.-R. and M.J.L. was supported by National Institutes of Health Grants R01 GM038464 and R15 GM079778, respectively.


The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1012363108/-/DCSupplemental.


1. Aguilera A. The connection between transcription and genomic instability. EMBO J. 2002;21:195–201. [PMC free article] [PubMed]
2. Aguilera A, Gómez-González B. Genome instability: A mechanistic view of its causes and consequences. Nat Rev Genet. 2008;9:204–217. [PubMed]
3. Beletskii A, Bhagwat AS. Transcription-induced mutations: Increase in C to T mutations in the nontranscribed strand during transcription in Escherichia coli. Proc Natl Acad Sci USA. 1996;93:13919–13924. [PMC free article] [PubMed]
4. García-Rubio M, Huertas P, González-Barrera S, Aguilera A. Recombinogenic effects of DNA-damaging agents are synergistically increased by transcription in Saccharomyces cerevisiae. New insights into transcription-associated recombination. Genetics. 2003;165:457–466. [PMC free article] [PubMed]
5. Huertas P, Aguilera A. Cotranscriptionally formed DNA:RNA hybrids mediate transcription elongation impairment and transcription-associated recombination. Mol Cell. 2003;12:711–721. [PubMed]
6. Kim N, Jinks-Robertson S. dUTP incorporation into genomic DNA is linked to transcription in yeast. Nature. 2009;459:1150–1153. [PMC free article] [PubMed]
7. Gangloff S, Lieber MR, Rothstein R. Transcription, topoisomerases and recombination. Experientia. 1994;50:261–269. [PubMed]
8. Morey NJ, Greene CN, Jinks-Robertson S. Genetic analysis of transcription-associated mutation in Saccharomyces cerevisiae. Genetics. 2000;154:109–120. [PMC free article] [PubMed]
9. Kim N, Abdulovic AL, Gealy R, Lippert MJ, Jinks-Robertson S. Transcription-associated mutagenesis in yeast is directly proportional to the level of gene expression and influenced by the direction of DNA replication. DNA Repair (Amst) 2007;6:1285–1296. [PMC free article] [PubMed]
10. Lippert MJ, Freedman JA, Barber MA, Jinks-Robertson S. Identification of a distinctive mutation spectrum associated with high levels of transcription in yeast. Mol Cell Biol. 2004;24:4801–4809. [PMC free article] [PubMed]
11. Wang JC. Cellular roles of DNA topoisomerases: A molecular perspective. Nat Rev Mol Cell Biol. 2002;3:430–440. [PubMed]
12. Pommier Y, et al. Repair of topoisomerase I-mediated DNA damage. Prog Nucleic Acid Res Mol Biol. 2006;81:179–229. [PMC free article] [PubMed]
13. Datta A, Jinks-Robertson S. Association of increased spontaneous mutation rates with high levels of transcription in yeast. Science. 1995;268:1616–1619. [PubMed]
14. Shinohara A, Ogawa H, Ogawa T. Rad51 protein involved in repair and recombination in S. cerevisiae is a RecA-like protein. Cell. 1992;69:457–470. [PubMed]
15. Nelson JR, Lawrence CW, Hinkle DC. Thymine-thymine dimer bypass by yeast DNA polymerase ζ Science. 1996;272:1646–1649. [PubMed]
16. Tanizawa A, Kohn KW, Pommier Y. Induction of cleavage in topoisomerase I c-DNA by topoisomerase I enzymes from calf thymus and wheat germ in the presence and absence of camptothecin. Nucleic Acids Res. 1993;21:5157–5166. [PMC free article] [PubMed]
17. Lehner K, Jinks-Robertson S. The mismatch repair system promotes DNA polymerase ζ-dependent translesion synthesis in yeast. Proc Natl Acad Sci USA. 2009;106:5749–5754. [PMC free article] [PubMed]
18. Lynn RM, Bjornsti MA, Caron PR, Wang JC. Peptide sequencing and site-directed mutagenesis identify tyrosine-727 as the active site tyrosine of Saccharomyces cerevisiae DNA topoisomerase I. Proc Natl Acad Sci USA. 1989;86:3559–3563. [PMC free article] [PubMed]
19. Pouliot JJ, Yao KC, Robertson CA, Nash HA. Yeast gene for a Tyr-DNA phosphodiesterase that repairs topoisomerase I complexes. Science. 1999;286:552–555. [PubMed]
20. Champoux JJ. DNA topoisomerases: Structure, function, and mechanism. Annu Rev Biochem. 2001;70:369–413. [PubMed]
21. Hsiang YH, Lihou MG, Liu LF. Arrest of replication forks by drug-stabilized topoisomerase I-DNA cleavable complexes as a mechanism of cell killing by camptothecin. Cancer Res. 1989;49:5077–5082. [PubMed]
22. Nitiss J, Wang JC. DNA topoisomerase-targeting antitumor drugs can be studied in yeast. Proc Natl Acad Sci USA. 1988;85:7501–7505. [PMC free article] [PubMed]
23. Daley JM, Palmbos PL, Wu D, Wilson TE. Nonhomologous end joining in yeast. Annu Rev Genet. 2005;39:431–451. [PubMed]
24. Henningfeld KA, Hecht SM. A model for topoisomerase I-mediated insertions and deletions with duplex DNA substrates containing branches, nicks, and gaps. Biochemistry. 1995;34:6120–6129. [PubMed]
25. Pouliot JJ, Robertson CA, Nash HA. Pathways for repair of topoisomerase I covalent complexes in Saccharomyces cerevisiae. Genes Cells. 2001;6:677–687. [PubMed]
26. Vance JR, Wilson TE. Repair of DNA strand breaks by the overlapping functions of lesion-specific and non-lesion-specific DNA 3′ phosphatases. Mol Cell Biol. 2001;21:7191–7198. [PMC free article] [PubMed]
27. Vance JR, Wilson TE. Yeast Tdp1 and Rad1-Rad10 function as redundant pathways for repairing Top1 replicative damage. Proc Natl Acad Sci USA. 2002;99:13669–13674. [PMC free article] [PubMed]
28. Liu C, Pouliot JJ, Nash HA. Repair of topoisomerase I covalent complexes in the absence of the tyrosyl-DNA phosphodiesterase Tdp1. Proc Natl Acad Sci USA. 2002;99:14970–14975. [PMC free article] [PubMed]
29. Deng C, Brown JA, You D, Brown JM. Multiple endonucleases function to repair covalent topoisomerase I complexes in Saccharomyces cerevisiae. Genetics. 2005;170:591–600. [PMC free article] [PubMed]
30. Takahashi T, Burguiere-Slezak G, Auffret Van der Kemp P, Boiteux S. Topoisomerase 1 provokes the formation of short deletions in repeated sequences upon high transcription in Saccharomyces cerevisiae. Proc Natl Acad Sci USA. 2011;108:692–697. [PMC free article] [PubMed]
31. Vance JR, Wilson TE. Uncoupling of 3′-phosphatase and 5′-kinase functions in budding yeast. Characterization of Saccharomyces cerevisiae DNA 3′-phosphatase (TPP1) J Biol Chem. 2001;276:15073–15081. [PubMed]
32. Nitiss JL, Nitiss KC, Rose A, Waltman JL. Overexpression of type I topoisomerases sensitizes yeast cells to DNA damage. J Biol Chem. 2001;276:26708–26714. [PubMed]
33. Pourquier P, et al. Effects of uracil incorporation, DNA mismatches, and abasic sites on cleavage and religation activities of mammalian topoisomerase I. J Biol Chem. 1997;272:7792–7796. [PubMed]
34. Wu J, Liu LF. Processing of topoisomerase I cleavable complexes into DNA damage by transcription. Nucleic Acids Res. 1997;25:4181–4186. [PMC free article] [PubMed]
35. Hashimoto H, Chatterjee S, Berger NA. Mutagenic activity of topoisomerase I inhibitors. Clin Cancer Res. 1995;1:369–376. [PubMed]
36. Balestrieri E, Zanier R, Degrassi F. Molecular characterisation of camptothecin-induced mutations at the hprt locus in Chinese hamster cells. Mutat Res. 2001;476:63–69. [PubMed]
37. Sikorski RS, Hieter P. A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics. 1989;122:19–27. [PMC free article] [PubMed]
38. Longtine MS, et al. Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast. 1998;14:953–961. [PubMed]
39. Lea DE, Coulson CA. The distribution of the numbers of mutants in bacterial populations. J Genet. 1949;49:264–285. [PubMed]
40. Spell RM, Jinks-Robertson S. Determination of mitotic recombination rates by fluctuation analysis in Saccharomyces cerevisiae. Methods Mol Biol. 2004;262:3–12. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...