WO2014127287A1 - Method for in vivo tergated mutagenesis - Google Patents

Abstract

The invention relates to methods for targeting mutagenesis to a particular genomic region as well as related compositions and kits thereof. In an embodiment, the invention relates to a method for in vivo targeted mutagenesis comprising selectively introducing localised DNA damage into a genome in vivo, targeting a pathway requiring long-range resection so as to form a single-stranded region during biasing repair and selectively mutating the single stranded region. In another embodiment, the invention relates to a method of contacting a cell having an integrated array of DNA binding sites with a fusion protein of an array specific DNA binding domain and a DNA mutator enzyme domain. In another embodiment, the invention relates to a method of contacting a cell with a fusion protein of a DNA binding domain and a DNA nuclease domain, and further biasing repair of localised DNA damage by targeting a pathway requiring long-range resection. In another embodiment, the invention relates to a method of in vivo targeted mutagenesis, comprising delivering a Cas9 nuclease and a synthetic gRNA, and contacting the cell with a mutator.

Description

METHOD FOR IN VIVO TARGETED MUTAGENESIS

Related Applications

This application claims priority under 35 U.S.C. § 119(e) to U.S. Provisional Application Serial No. 61/764,691, entitled "METHOD FOR IN VIVO TARGETED MUTAGENESIS," filed on February 14, 2013, which is herein incorporated by reference in its entirety.

GOVERNMENT SUPPORT

This invention was made with government support under Grant No. DGE0645960 awarded by the National Science Foundation and under Grant No. P30-ES002109 from the

National Institute of Environmental Health Sciences. The government has certain rights in this invention.

FIELD OF THE INVENTION

The present invention relates generally to methods for targeting mutagenesis to a particular genomic region as well as related compositions and kits thereof.

BACKGROUND OF INVENTION

Clustered DNA damage, where multiple lesions are present within a few helical turns of DNA, is a signature of both exogenous (ie. ionizing radiation) and endogenous (i.e. reactive oxygen/nitrogen) damage sources as well as chemo- and radio-therapeutic agents. Processing of lesions through base and nucleotide excision repair pathways often results in clusters of abasic sites which can ultimately lead to single- or double- stranded breaks (SSB or DSB) in DNA [1, 2]. Three functional outcomes of this damage are 1) error- free repair, 2) mutagenic repair, and 3) cell death, with the likelihood of the latter outcomes increasing with the severity of the damage. Radiotherapy is a double-edged sword. When a tumor is exposed to high levels of ionizing radiation tumor cells are killed, but surrounding tissue exposed to slightly lower doses may experience less severe clustered DNA damage that could lead to mutation, genetic instability, and tumorigenesis [3] . In addition, intermediate doses could lead to both clustered and isolated damage; having both types of damage present may affect processing during repair.

The repair of DSBs occurs via homology and non-homology mediated processes, an understanding of which has in large part derived from elegant studies that introduced well- defined DSBs using site-specific endonucleases [4, 5]. Mutagenic repair of DSBs can lead to not only gross chromosomal rearrangements (GCRs) and chromosomal aberrations, but also point mutations (PMs) in regions adjacent to the break site [6]. While repair processes involving homologous recombination (HR) are traditionally seen as error-free [5, 6], PMs arising from DSB repair have been reported in yeast [7-9] . HR-mediated DSB repair via gene conversion requires resection of DNA ends, followed by extension of the ends using a homologous donor template. When the donor has limited homology, ends can switch between multiple donor templates, a potentially mutagenic process [10]. After annealing of the extended end, resynthesis of remaining ssDNA can also be mutagenic. If unrepaired DNA lesions accumulate on ssDNA exposed after resection, synthesis past these lesions leads to PMs and requires the TLS polymerase Pol ζ [8, 9]. Finally, DSB repair by break-induced replication (BIR) has also been shown to result in elevated frameshift mutations during synthesis of the donor template, as Pol δ fidelity decreases because of increased dNTP pools and inefficient mismatch repair [11].

DSB repair in bacteria can also be highly mutagenic and is regulated by stress.

Activation of both the DNA damage-specific SOS pathway and the general RpoS stress pathway leads to error-prone resynthesis of resected DSB ends by DinB and Pol V polymerases [12, 13]. Both SOS and RpoS increases DinB expression; additionally, RpoS increases expression of other low-fidelity polymerases and licenses DinB by an unknown mechanism [14]. In yeast, stress-induced mutagenesis has been suggested to occur in the context of DSB repair via an error-prone NHEJ pathway [15]), although parallels exist between the yeast and bacterial error- prone HR repair pathways [16].

SUMMARY OF INVENTION

In some aspects the invention is a method for in vivo targeted mutagenesis by selectively introducing localized DNA damage in a preselected region of an organism' s, optionally a eukaryote's, DNA in vivo, biasing repair of the localized DNA damage by targeting a pathway requiring long-range resectioning of the localized DNA damage, wherein the DNA forms a single stranded region during the biasing repair, and selectively mutating the single stranded region to cause targeted mutagenesis. In some embodiments the localized DNA damage is a double stranded break (DSB). In other aspects the invention is a method of contacting a cell with a fusion protein, wherein the fusion protein is an array specific DNA binding domain and a DNA mutator enzyme domain, wherein the cell has an integrated array of DNA binding sites to which the DNA binding domain is capable of binding.

In yet other aspects the invention is a method of contacting a cell with a fusion protein, wherein the fusion protein is a DNA binding domain and a DNA nuclease domain in order to produce a localized DNA damage, and biasing repair of the localized DNA damage by targeting a pathway requiring long-range resectioning of the localized DNA damage. In some embodiments the localized DNA damage is a double stranded break (DSB).

In some embodiments the localized DNA damage is introduced by a DNA mutator enzyme domain. The DNA mutator enzyme in some embodiments is a DNA glycosylase, such as 3-methyladenine glycosylase Maglp (Maglp) or UDP. In other embodiments the DNA mutator enzyme is a DNA nuclease, such as Fokl.

The DNA binding domain in some embodiments is a TAL binding domain or a zinc finger binding domain. In some instances a DNA binding domain is fused to a nuclease, for example as a TALEN, or multiple TALENs.

The methods may involve contacting the cell with a compound that elicits DNA damage checkpoint activation. In some embodiments the compound that elicits DNA damage checkpoint activation is a chemical checkpoint activator such as MMS or BSO. In other embodiments the compound that elicits DNA damage checkpoint activation is an enzymatic checkpoint activator such as Mag 1.

The methods involve in some embodiments an array specific DNA binding domain, such as, for example, a tetR or a (sc)tetR. In some embodiments the integrated array of DNA binding sites is a tetO array such as an 85x or 240x tetO array.

The method may be performed on any type of cell. In some embodiments the cell is a yeast cell. In other embodiments the cell is a mammalian cell or a plant cell. In other embodiments the cell is not a cell of human B lymphocyte lineage.

The methods of the invention may also involve downregulating non-HR processes in the cell. The downregulation of non-HR processes in the cell in some embodiments involves contacting the cell with an inhibitory nucleic acid.

In other embodiments the cell is contacted with a recombinant nucleic acid capable of expressing Magi. In other aspects the invention is a fusion protein of a DNA binding domain and a DNA mutator enzyme domain, wherein the DNA mutator enzyme is not a deaminase protein. The fusion protein in some embodiments is sctetR-Fokl or Magl-sctetR.

In other aspects the invention is an isolated nucleic acid sequence that encodes the fusion proteins described herein, an expression vector, comprising the isolated nucleic acid or a host cell, comprising the expression vector.

A plant having a mutated germline made according to any of the methods described herein is also provided.

Each of the limitations of the invention can encompass various embodiments of the invention. It is, therefore, anticipated that each of the limitations of the invention involving any one element or combinations of elements can be included in each aspect of the invention. This invention is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the drawings. The invention is capable of other embodiments and of being practiced or of being carried out in various ways. Also, the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of "including," "comprising," or "having," "containing", "involving", and variations thereof herein, is meant to encompass the items listed thereafter and equivalents thereof as well as additional items. BRIEF DESCRIPTION OF DRAWINGS

The accompanying drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component that is illustrated in various figures is represented by a like numeral. For purposes of clarity, not every component may be labeled in every drawing. In the drawings: FIG. 1 Is an illustration showing targeted mutagenesis using DNA binding domain fusions. Plasmid containing a protein fusion of the DNA glycosylase MAGI to single-chain (sc)tetR under galactose-inducible control is introduced into cells containing a 240x tetO binding site array. Targeted and untargeted mutation rates were tested using the URA3 and CAN1 markers, respectively. Deletion of APN1 is one example of a DNA repair pathway modification that may further increase targeted mutation rates. FIG. 2 Shows how targeted mutators increase the mutation rate in a 20 kb area surrounding the tetO array. A) The Magl-sctetR fusion is expressed from a galactose-inducible promoter on a centromeric plasmid present in cells containing a 240x tetO array integrated on the right arm of chromosome I. The mutation rate marker KIURA3 is introduced at various positions near the array (one instance per strain). B) In cells with the tetO array and expressing Magl-sctetR, targeted mutation rates at 0.3 kb are elevated 800-fold as measured by fluctuation assays, while rates at the CAN1 marker on chromosome V do not change. (C) This increase in mutation rate persists for at least 10 kb on either side of the array, with selection for HIS 3 (diamonds) decreasing the mutation rate slightly and addition of dox (squares) eliminating targeted mutagenesis completely. Labels on data points report the ability to PCR KIURA3 from a particular mutant, PCR-i-(total). Error bars represent 95% confidence limits.

FIG. 3 Shows that mutations from Magl-sctetR and sctetR-Fokl (TaGTEAM)are created during HR repair of targeted damage. A) TaGTEAM in a strain carrying a CFP-tagged version of Rad52p shows that damage at the array is repaired through HR. Error bars are bootstrapped 95% confidence limits. B) A model for mutagenesis through HR that generates rearrangements due to short repetitive sequences or point mutations through resection, DNA damage, and pol ζ recruitment during resynthesis. C) Knockout mutants of pathway components in B) demonstrate that targeted mutagenesis depends on HR (RAD52 )and that point mutations (dominant under HIS 3 selection) depend on REV3 and SGSI +EX01. Error bars represent 95% confidence limits.

FIG. 4 Shows how cellular DNA damage context explains the difference between the repair of breaks generated by each mutator. Global DNA damage redirects mutagenic repair of sctetR-Fokl- induced breaks towards HRdependent point mutations via checkpoint activation and DNA lesions. Mutation rates generated by sctetR-Fokl expression in WT, Pol ζ-deficient (rev3), and checkpoint-deficient (smll ddc2) strains were measured and compared to those in the presence of co-expressed Maglp, MMS, or HU. A). In the absence of selection for HIS3, co- expression of Maglp with sctetR-Fokl makes checkpoint- and Pol ζ- dependent point mutagenesis the dominant mutagenic outcome, as indicated by scoring of mutants for a HIS+ and/or PCR+ phenotype (listed above bars). HU, on the other hand, decreases HR-independent rearrangements without creating point mutations. In the absence of Magi activity, loss of checkpoint activation leads to very high (>10^-4 cell^"1 gen^"1) mutation rates that correspond to rearrangements. B) HIS3 selection reveals Pol ζ-dependent point mutations generated by the addition of MMS. In every case observed, his- mutants were never PCR+. Addition of HU to smll ddc2 strains eliminates growth, preventing measurement of the mutation rate. Error bars represent 95% confidence limits.

FIG. 5 Shows two conditions that switch the primary mutagenic repair outcome of DSBs from GCRs to PMs. Targeted DSBs generated by Fokl lead to HR-independent rearrangements and not point mutations. A) Expression of sctetR-Fokl in the same strain background as Magl- sctetR (B) leads to a similar (620-fold) increase in targeted mutation rates without any increase in background mutation rates. C) The sctetR-Fokl distance dependence is asymmetric, selection for HIS3 leads to a more severe drop in mutation rate, and very few PCR+ mutants are generated as compared to Magi -sctetR. D) Rad52-CFP repair foci show that sctetR-Fokl damage is repaired by HR in roughly the same fraction of cells as Magi -sctetR, but E) targeted

mutagenesis is not RAD52 -dependent , and even under selection for HIS 3 there is no REV3- dependence on. Error bars as in Fig. 3.

FIG. 6 Shows how the Magi -sctetR fusion protein is able to bind tetO and mutate DNA. (A) MAGI overexpression in WT cells leads to an increase in the background mutation rate at CAN1. The MAGI -sctetR fusion has decreased but significant mutator activity, as evidenced by the increased mutation rate in an αρηΙΔ. Surprisingly, MAGI overexpression in an αρηΙΔ does not lead to a measurable increase in the mutation rate. This is (B) due to a severe growth defect, which is (C) relieved specifically upon fusion to sctetR and does not depend on sctetR' s ability to bind DNA (+dox growth curve). (D) The reduction in mutator function upon fusion to sctetR could in part be due to decreased expression levels as measured by flow cytometry on YFP- tagged mutators. (E) The AG-sctetR fusion protein retains the ability to bind tetO as measured by fluorescence knockdown from a tet-repressible promoter. Error bars on mutation rate represent 95% C.I. Error bars on fluorescence knockdown and growth represent the range of values observed in triplicate experiments.

FIG. 7 Shows how inter and intra-chromosomal repetitive homologous sequences lead to deletions. Various repetitive homologous sequences introduced during strain construction— 17 bp, 18 bp, 201 bp, 430 bp— can mediate different HR dependent deletions of the mutation rate marker K1URA3 depending on its position. At 0.3 kb there are three possible deletions, only one of which leads to simultaneous deletion of the HIS3 marker. At all other positions, there is only one possible deletion, and it always results in simultaneous deletion of the HIS3 marker.

FIG. 8 A model for the mutagenic fate of targeted damage generated by Magl-sctetR or sctetRFokl. Magl-sctetR and sctetR-Fokl both generate lesions that lead to DSBs, but the mutagenic repair outcome depends on two conditions that switch repair: 1) checkpoint activation and 2) base pair damage. A) In cells expressing Magl-sctetR both conditions are met, leading to high rate point mutagenesis and minimal HR independent rearrangement. B) sctetR-Fokl expressing cells do not activate the DNA damage checkpoint to the same extent or experience base pair damage and the primary mutagenic event is HR-independent rearrangements. Co- expression of untargeted Maglp with sctetR-Fokl (C) or addition of 0.001% MMS (D) or 4.5 mg/mL HU (E), demonstrates that the transition in primary mutagenic outcome from HR- independent rearrangements to HR and REVJ-dependent point mutations occurs only when both conditions are met The percentages reported here are summarized in Table 2.

FIG. 9 Illustrates mutation rates using two different mutator constructs in combination with chloroacetaldehyde (CAA) treatment.

FIG. 10 Illustrates the use of constitutive expression of Magl-sctetR from a strong, commonly used promoter to target mutagenesis in a variety of carbon sources in prototrophic lab (S288c) and industrial (Ethanol Red) strains of yeast. A) Targeted mutagenesis on a plasmid at the gain of function marker ade2- l that reverts through modification of a stop codon shows that TaGTEAM of plasmid-borne genes is possible in yeast. B) Constitutive expression of Magl- sctetR allows targeted mutagenesis in a variety of carbon sources in prototrophic lab (S288c) and industrial (Ethanol Red) strains of yeast.

FIG. 11: Measurement of background mutation rates at CAN1 by fluctuation analysis. Rates (left) are calculated by fitting the number of mutant cells from 12 parallel cultures to the Luria-Delbruck distribution using maximum likelihood estimation. Comparison of empirical data to the best-fit cumulative distribution (right). Error bars represent 95% c.i. (methods).

FIG. 12: CDG-sctetR does not retain ability to mutate DNA. (A) CDG but not CDG- sctetR increases the background mutation rate in αρηΙΔ. (B) Expression of a nuclear localization signal (NLS)-tagged CDG-vYFP fusion was measured by fluorescence microscopy. Histograms represent cellular autofluorescence- subtracted YFP expression in arbitrary units (AU) as measured by fluorescence microscopy. Expression of CDG-YFP is significantly lower than both NLS-vYFP and A i-vYFP (Fig. 12C), possibly explaining its lack of activity. Error bars represent 95% c.i.

FIG. 13: Localization of tetR-YFP and YFP foci observation confirms 240x tetO array presence in point mutants. Transformation of a plasmid delivering a methionine-inducible fusion of tetR to YFP shows that PCR+ mutants created in the absence of selection for HIS3 retain the array while PCR- mutants do not. Under selection, all PCR+ and most PCR- mutants retain the array, consistent with a KIURA3 deletion by repetitive homology that preserves the HIS3 marker (see Fig. 7).

FIG. 14: Cell cycle distributions show importance of DNA damage checkpoint activation in DSB repair fate. Compared to sctetR-Fokl, Magl-sctetR expression increases the fraction of cells with 2C DNA content as determined by flow cytometric analysis of

exponentially growing cells stained with SYTOX green. This increase is indicative of the DNA damage checkpoint activation because it is eliminated in checkpoint-deficient (smll ddc2) strains. Co-expression of Maglp causes increased checkpoint activation as compared to expression of sctetR-Fokl alone. Checkpoint-deficient strains co-expressing Maglp grow significantly slower than other strains, explaining why the number of cells with 2C DNA content increases in this case.

FIG. 15: Titration of MMS level reveals a plateau in mutation rate. A) Overnight growth of cells in various levels of MMS compared to growth without MMS. B) Mutation rates in cells expressing sctetR-Fokl reach a maximum at 0.003% MMS. Selection for HIS3 reveals that the majority of mutations at this level of MMS are point mutations.

FIG. 17: Nocodazole arrest increases the targeted PM rate. sctetR-Fokl expressing cells were arrested. Targeted mutation rates increased 15-fold under HIS3 selection. PM rates peaked at 4 hours, while mutations generated in the absence of HIS3 selection increased further at 6 hours. "0 hrs" represents controls that were transferred directly from pre-growth to grow-out media. Mutation rates in the absence of the array and presence of MMS were also elevated compared to the "Ohrs" case, which is >1000-fold lower in the absence of the array and not presented here. Error bars represent 95% c.i.

FIG. 18: Nocodazole arrest increases the fraction of cells with Rad52-CFP foci, but not the chance that a G2 cell experiences a TaGTEAM induced break. sctetR-Fokl expressing cells with the 240x tetO array under arrest experienced a 3-fold higher fraction of cells with Rad52- CFP foci. These foci were predominantly in G2 cells, the fraction of which was also enriched 3- fold in the overall population, suggesting that arrest does not increase the chance that a G2 cell experiences a break. In addition longer arrest times did not lead to higher fractions of cells with foci. This is in contrast to cells without the tetO array that experienced increased foci generation over time in arrest and a dramatic increase in the fraction of foci that were in M cells after 6 hours of arrest. Error bars represent 95% c.i. - SI - FIG. 19: Converting targeted genome modification methods to targeted mutagenesis methods. (A) Using various nuclease targeting strategies, DSBs can be introduced at specific locations within a genome. Sub-lethal doses of particular enzymes or chemicals are added to introduce lesions. (B) Normally these lesions are repaired, but if they occur near a DSB, they are converted to point mutations at high efficiency.

FIG. 20: Design of a system to evolve bZIP interactions. A) Fusions of each bZIP domain, a fluorescent protein, and either the Lex A DNA binding domain or VP 16 will be expressed under methionine control on a centromeric plasmid next to an 85x array. The ade2-l marker will also be placed on this plasmid to allow measurement of mutation rates periodically. B) bZIP domains which have interactions ranging from high affinity to repulsive will be evolved. C) Galactose-controlled mutators will be integrated at the LEU2 locus and two different markers at different locations will link selection or screening to the strength of the bZIP interaction.

Table 1: Magl-sctetR generates a broad spectrum of point mutations

Table 2: Different types of loss function mutations are determined by genotype and phenotype

Table 3: Cell cycle distribution and growth rate of mutator expressing strains.

Table 4: Strain List

Table 5: Plasmid List

Table 6: Primer List

Table 7: Mutation spectrum with CAA

DETAILED DESCRIPTION OF INVENTION

Localized DNA damage, such as double-strand breaks (DSBs) and double strand ends, are a source of DNA damage that can be lethal. Cells have repair mechanisms that recognize and repair some of the localized DNA damage. However, the repair of localized DNA damage may be achieved in an error- free manner or an error-prone manner that results in DNA mutations. The nature and context of the localized DNA damage influences the repair outcome.

A system was developed according to the invention that targets different enzymes to generate clustered DNA damage at a well-defined genomic location. It was found that mutagenic repair of this localized DNA damage can lead to either gross chromosomal rearrangements (GCRs) or point mutations (PMs) in a surrounding 20 kb region. GCRs are largely due to Ku- dependent repair of the break. However, additional stress can increase Mecl/ATR checkpoint activation, preventing Ku-dependent repair. If this stress also introduces DNA lesions, it results in long range point mutagenesis surrounding the DSB during repolymerization of resectioned ends by the trans-lesion synthesis (TLS) polymerase Pol ζ. Even low levels of global DNA damage can shift the mutagenic repair from GCRs to PMs. As most GCRs are lethal, this stress- induced switch increases viability at the cost of PMs in the presence of clustered DNA damage, which might aid in adaptation of yeast but promote tumorigenesis if present in multicellular organisms.

The Examples included herein describe a system that was developed according to the invention to damage DNA at a well-defined genomic location in organisms, such as S.

cerevisiae, ultimately leading to localized DNA damage whose mutagenic repair elevates mutations at least a 10²-10³-fold within an extended region -10 kb adjacent to the break. The localized DNA damage that was introduced, mimics natural contexts by originating from either clusters of abasic sites and/or at low enough rates such that multiple cleavage events don't always occur in one cell cycle period. Using this strategy, a switch was identified in the mutagenic repair of the localized DNA damage from GCRs to long range PMs that depends on both general stress signals that elicit checkpoint delay in S/G2 (Mecl/ATR) and low levels of genome- wide DNA damage stress.

The in vivo targeted mutagenesis methods of the invention may also be referred to in some embodiments as TArgeting Glycosylases To Embedded Arrays for Mutagenesis

(TaGTEAM). For example, it is shown herein that by fusing the yeast 3-mefhyladenine DNA glycosylase MAGI to a tetR DNA binding domain, it was possible to achieve elevated mutation rates > 800 fold in a specific -20 kb region of the genome or on a plasmid that contains an array of tetO sites, with no change in background mutation rates within the genome. A wide spectrum of transitions, transversions, and single base deletions were observed. We provide evidence that TaGTEAM generated point mutations occur through error-prone homologous recombination (HR) and depend on resectioning and the error prone polymerase Pol ζ. Additionally it is shown that HR is error-prone in this context because of DNA damage checkpoint activation and base pair lesions. This knowledge is used to shift the primary mutagenic outcome of targeted endonuclease breaks from HR-independent rearrangements to HR-dependent point mutations. The ability to switch repair in this way presents the opportunity to use targeted endonucleases in diverse organisms for in vivo targeted mutagenesis.

Thus, knowledge of the underlying mechanism behind mutations created by TaGTEAM should allow for new ways of using Magl-sctetR, sctetR-Fokl, and other targeted mutators. Targeted mutagenesis relies on homologous recombination, long range resectioning, and base pair damage. This invention includes methods to increase targeted mutation rates by altering the cellular context in such a way that these processes are favored. In addition to increased base pair damage by chemical agents like MMS and CAA, targeted mutagenesis can be increased by altering the growth and or mutator expression timing of the population being mutated.

The methods can be accomplished, for instance, by growing up a large population of cells without expressing the mutator and then arresting them all at the G2/M cell cycle checkpoint while turning on expression of the mutator. HR proteins are upregulated in these post-replication cells, and, because the cell cycle is stopped, there is sufficient time for long range resectioning to occur. More cells on average experience a clustered damage event that leads to HR because there is more time for clustered damage to occur. Arrest at the G2/M DNA damage checkpoint may be accomplished, for example, using chemicals like nocodazole and mutants in cell cycle kinases such as CDC 15 or CHK1 in yeast. This includes arrest at the intra- S checkpoint achieved by chemicals such as hydroxyurea or camptothecin.

Another way to alter the cellular context is to grow the cells in concentrations of the cell cycle arresting chemicals such as nocodazole that do not lead to full cell cycle arrest but instead cause only an extension of the S and/or G2 phases of the cell cycle. The mutator construct is expressed continuously, and the mutation rate is increased because cells spend longer in a cell cycle phase were HR is the dominant repair pathway. Also, a clustered damage event and long range resectioning are more likely to occur because of slow cell cycle progression.

Using the methodology of the invention it is now possible to selectively mutagenizing any region of the genome in a wide variety of cell types by simply expressing a sequence- specific nuclease and treating with appropriate chemical or enzymatic mutagens (dictates the type of mutation). Note once a cell is engineered to express the targeted nucleases the mutagenic, diversity-producing process is in vivo and does not require transformation or transfection. Still, challenges remain to increase mutagenic efficiency. We can improve the targeted mutation rate to -10^"4 bp^"1 gen^"1 by simply arresting the cells in the G2 phase of the cell cycle when HR is naturally upregulated.

The methods have applicability in a wide variety of system, including for instance, human embryonic stem cells or plant cells where the technology can be used for in vivo mutagenesis to improve the reprogramming and differentiation efficiency of stem cells, or improve the resistance of plants to pathogens.

The methods of the invention, through sequential mutagenesis and selection, represent a novel method for directed evolution in multiple organisms including yeast and eukaryotes. In the methods mutagenesis occurs continuously in vivo without the need for rounds of genetic transformation, which is a great improvement over existing technology. Due to yeast's industrial relevance, TaGTEAM is an important step toward the in vivo directed evolution of relevant multigenic cellular phenotypes including metabolic pathways, synthetic regulatory networks, and tolerance to chemicals of interest present in industrial fermentation. To this end, we have demonstrated the ability of TaGTEAM to function in industrially relevant strains under constitutive control.

The methods can also be used for in vivo targeted mutagenesis in higher organisms like plants, mice, or human cells. Based on the difficulty of transforming these organisms, the ability to do targeted mutagenesis make it feasible to perform directed evolution of medically and industrially relevant phenotypes like stem cell differentiation or drought resistance in crop plants. Even though imprecise NHEJ is the dominant DSB repair pathway in higher eukaryotes, consequential mutations like those found in breast cancer tumors contain a signature consistent with their generation by long range error-prone HR. Targeted endonucleases allow for the generation of DSBs that are repaired by imprecise NHEJ. In contrast, the methods of the invention involve DSB by error-prone HR via long-range resectioning. Importantly, the Cas9 endonuclease can be expressed with multiple guide RNAs, allowing for the simultaneous generation of multiple breaks within any genomic region.

Two different approaches are useful for downregulating imprecise NHEJ without the requirement of generating knockout mutants of genes in the NHEJ pathway. The first relies on RNA interference of important genes in the NHEJ pathway. This approach has been

demonstrated in A. thaliana, where RNAi to XRCC4 increased the ability to integrate foreign genetic material through HR. XRCC4 mediates the interaction between Ku bound DNA ends and LiglV, the ligase that does the joining in NHEJ. Without it MRN eventually kicks off the Ku heterodimer signaling the start of resectioning. Another approach involves the use of small molecules that selectively inhibit the kinases, DNA-PKcs, which promote NHEJ at a DSB. A survey of these molecules in Chinese hamster ovary cells found several (NU7206 or NU7441) that inhibit DSB repair to the same extent as a Ku-/- or DNA-PKc-/- cell line, suggesting that they inhibit NHEJ.

Base damaging agents and promotion of resectioning make HR capable of generating long range, high rate PMs. After making HR the dominant repair pathway of targeted nuclease induced breaks, the next step is ensuring that those breaks lead to long range PMs at high rates. As discussed here, the addition of small amounts of CAA or MMS can lead to a dramatic increase in the point mutation rate in yeast cells undergoing TaGTEAM. CAA is slightly more specific to damage of ssDNA, decreasing the off target effects of its addition. An additional chemical that may allow further increase in ssDNA damage without dsDNA damage or cellular toxicitiy is sodium bisulfite. Sodium bisulfite leads to conversion of cytosine to uracil, and has been shown to generate hypermutagenesis in exposed ssDNA in yeast (Chan et al., 2012).

An additional important step in increasing the targeted mutation rate is the promotion of resectioning and removal of the growth bias incurred by mutants as opposed to non-mutants. As discovered in yeast experiencing TaGTEAM, the effect of cell cycle time and strain variation can have an effect on the PM rate. For instance, as described herein in more detail, G2 arrest with nocodazole leads to increased mutation rates. This increase may be due to the marked preference of HR in this cell cycle stage. Because 4 hour arrest leads to higher HIS+ point mutations than 6 hour arrest, further stalling past 4 hours or so does not cause promoter resectioning (although up to 4 hours it may). Alternatively the increase in mutation rate due to arrest may arise because arrest eliminates any growth competition from non-mutants. For example, if a mutating cell stalls in a rapidly growing culture while it is repairing itself, then there will be fewer mutants when that culture reaches saturation compared to if the repair time was quick. Rad52-FP foci can be used to determine the break rate, and this can be compared to the mutation rate in strains experiencing adequate base damage. If it appears that culture arrest or the promotion of resectioning would be helpful in generating more mutants, similar methods can be used. Cell cycle arrest can be carried out in mammalian cells through the addition of camptothecin (intra-S checkpoint) or nocodazole (G2/M checkpoint). Promotion of resectioning is possible in mammalian cells through over-expression of the mammalian SGS1 and EXOl counterparts BLM and Exol. In vivo targeted mutagenesis in higher organisms requires generating a targeted break or breaks, funneling that break to HR and not NHEJ, and then making sure HR is error-prone at as high a rate as possible. The ability to introduce these pieces either chemically or through transient expression on a transfected plasmid means that in vivo targeted mutagenesis can be carried out for a few generations before cells naturally lose the plasmid. Similar plasmids can be used to attempt targeted mutagenesis in various cell types within a particular kingdom.

Leveraging insights from the promotion of error-prone HR in yeast to allow for in vivo targeted mutagenesis in plants or mammals adds a third functionality to the up and coming genome editing technology of targeted endonucleases. Instead of just generating knockouts or aiding in integration of new genetic material, these endonucleases can be harnessed for targeted generation of genetic diversity, allowing for the directed evolution of phenotypes within these organisms that have tremendous medical and industrial relevance.

Thus, the invention in some aspects is a method for targeting mutagenesis to a particular genomic region in vivo. The method involves selectively introducing localized DNA damage in a preselected region of a mammalian organism' s DNA in vivo, biasing repair of the break by targeting a pathway requiring long-range resectioning of the localized DNA damage, wherein the DNA forms a single stranded region during the biasing repair, and selectively mutating the single stranded region to cause targeted mutagenesis.

DNA damage can be introduced in a preselected region of a mammalian organism' s DNA in vivo or cellular DNA through the use of a protein, referred to herein as DNA mutator enzyme domain. The DNA damage can lead to a DSB or double strand end (mutation in dsDNA). Preferably the DNA mutator enzyme domain is directed to specific sites in the DNA using a targeting molecule. For instance, in one embodiment the DNA mutator enzyme domain may be fused to a protein that binds to specific DNA sites. Such a protein is referred to herein as a DNA binding domain.

A "DNA mutator enzyme domain" as used herein refers to an enzyme or active fragment thereof that causes the introduction of DNA damage that may result in a DSB in double stranded DNA. The DNA mutator enzyme is directed to bind to specific DNA sequences or motifs through it' s interaction with a DNA binding domain. The "DNA binding domain" as used herein is a protein or fragment thereof that specifically interacts with DNA sequences or motifs. The DNA mutator enzyme may be fused to the DNA binding domain in the form of a fusion protein. For instance, such a fusion protein could be prepared using recombinant techniques, as described below, and expressed in a cell, i.e. the cell may be contacted with a recombinant nucleic acid capable of expressing fusion protein.

In some instances the DNA mutator enzyme is a DNA glycosylase. DNA glycosylases include but are not limited to 3-methyladenine glycosylase Maglp (Maglp) and uracil DNA glycosylases. DNA glycosylases are a family of enzymes involved in base excision repair, the mechanism by which damaged bases in DNA are removed and replaced. DNA glycosylases remove a damaged base while leaving the sugar-phosphate backbone intact, creating an abasic nucleotide (apurinic/apyrimidinic site). Monofunctional glycosylases have glycosylase activity and bifunctional glycosylases also have the ability to cleave the phosphodiester bond of DNA, creating a single- strand break without the need for an endonuclease.

Glycosylases consist of four families, the UDG, AAG, MutM/Fpg and HhH-GPD families. The UDG and AAG families contain small, compact glycosylases, whereas the MutM/Fpg and HhH-GPD families comprise larger enzymes with multiple domains. Uracil- DNA glycosylase (UDG) excises uracil residues from DNA by cleaving the N-glycosylic bond, initiating the base excision repair pathway. 3-methyl-adenine DNA glycosylase is involved in protecting DNA against alkylating agents and initiates base excision repair by removing damaged bases to create abasic sites.

3 -Methyl Adenine DNA Glycosylase 2, for instance has the following amino acid sequence, (SEQ ID NO: 1):

1 MKLKREYDEL IKADAVKEIA KELGSRPLEV ALPEKYIARH EEKFNMACEH

51 ILEKDPSLFP ILKNNEFTLY LKETQVPNTL EDYFIRLAST ILSQQISGQA

101 AESIKARVVS LYGGAFPDYK ILFEDFKDPA KCAEIAKCGL SKRKMIYLES

151 LAVYFTEKYK DIEKLFGQKD NDEEVIESLV TNVKGIGPWS AKMFLISGL 201 RMDVFAPEDL GIARGFSKYL SDKPELEKEL MRERKVVK S KIKHK YNWK

251 IYDDDIMEKC SETFSPYRSV FMFILWRLAS TNTDAMMKAE ENFVKS

The DNA mutator enzyme may also be a DNA nuclease domain. A DNA nuclease domain is an enzymatically active protein or fragment thereof that causes DNA cleavage resulting in a DSB. DNA nucleases include but are not limited to Fokl including monomeric Fokl and scFokl. Fokl is a type IIS restriction endonuclease having an N-terminal DNA-binding domain and a C-terminal non-specific DNA cleavage (endonuclease) domain. The DNA-binding domain of Fokl recognizes a DNA site at the 5'-GGATG-3': 5'-CATCC-3'. The endonuclease domain is formed by the parallel helices 4 and 5 and two loops PI and P2 of the cleavage domain. Once bound to endonuclease domain will cleave the DNA at a specific site

downstream and upstream of the recognition site. Monomeric Fokl can dimerize with another monomer copy to form an active dimer complex. scFokl refers to single chain Fokl. Single chain Fokl is two copies of monomeric Fokl covalently attached with a peptide linker. DNA nucleases and other mutation enzyme domains may be fused with DNA binding domains to produce the DSBs in the target DNA of the methods of the invention. DNA binding domains include, for example, an array specific DNA binding domain or a site specific DNA binding domain. Site specific DNA binding domains include but are not limited to a TAL (Transcription Activator-Like Effector) or a zinc finger binding domain. Examples of DNA-binding domains fused to DNA nucleases include but are not limited to TALEN and multiple TALENs. Transcription Activator-Like Effector Nucleases (TALENs) are artificial restriction enzymes generated by fusing the TAL effector DNA binding domain to a DNA enzyme domain. TAL proteins are produced by bacteria and include a highly conserved 33-34 amino acid DNA binding domain sequence. The 12th and 13th amino acids of this conserved region are highly variable (Repeat Variable Diresidue) and show a strong correlation with specific nucleotide recognition. Taking advantage of the relationship between amino acid sequence and DNA recognition has allowed for the engineering of Transcription activator- like effectors (TALEs) to bind to a wide variety of DNA sequences.

The original TALEN chimera were prepared using the wild-type Fokl endonuclease domain. However, TALEN also include chimera made from Fokl endonuclease domain variants with mutations designed to improve cleavage specificity and cleavage activity. The Fokl domain functions as a dimer and as such is typically composed of two constructs with unique DNA binding domains for sites in the target genome with proper orientation and spacing. Both the number of amino acid residues between the TALE DNA binding domain and the Fokl cleavage domain and the number of bases between the two individual TALEN binding sites may affect the levels of activity.

A zinc finger is another type of DNA binding domain useful according to the invention. A zinc finger is a small protein structural motif that is characterized by the coordination of one or more zinc ions in order to stabilize the fold. A large variety of zinc finger proteins exist. However, the majority of these proteins typically function as interaction modules that bind DNA, RNA, proteins, or other molecules. The zinc finger protein domains useful in the invention are any zinc finger domains that bind DNA.

The proteins of the CRISPR system are examples of other DNA -binding and DNA nuclease domains. For instance, Cas9 nuclease produces double strand breaks when complexed with an appropriate guide RNA, forming a DNA specific complex. dCas9, may also be useful in the invention for helping target a functional nuclease to DNA sites. dCas9 is a catalytically dead analog of Cas9 that has a D10A and H840A mutation, but other versions of dCas9 are also envisioned. In some embodiments Cas9 or dCas9 is linked to Magi. Thus, the invention contemplates fusion proteins of Cas9-Magl and dCas9/Mag 1.

Various protein engineering techniques can used to alter the DNA-binding specificity of zinc fingers and tandem repeats of such engineered zinc fingers can be used to target desired genomic DNA sequences. Fusing a second protein domain such as a transcriptional activator or repressor to an array of engineered zinc fingers that bind near the promoter of a given gene can be used to alter the transcription of that gene. Typical engineered zinc finger arrays have between 3 and 6 individual zinc finger motifs and bind target sites ranging from 9 basepairs to 18 basepairs in length. The zinc finger arrays may be fused to a DNA mutator enzyme domain (for example the endonuclease domain of Fokl) to generate zinc finger nucleases. These constructs can be used to target the DSB to a specific genomic locus to which the zinc finger protein is designed to interact. Several engineered zinc finger arrays are based on the zinc finger domain of the murine transcription factor Zif268 or the human transcription factor SP1. Zif268 has three individual zinc finger motifs that collectively bind a 9 bp sequence with high affinity.

In some instances the DNA mutator enzyme domain and the DNA binding domain are a fusion protein of TALEN fused to dimeric scFokl. In other instances multiple TALENs can be expressed to target multiple genomic regions. The DNA binding domain in some embodiments is an array specific DNA binding domain. The methods of the invention can be achieved using an array specific DNA binding domain by introducing an array of DNA binding sites into the cell or organism and targeting the array with a fusion protein that includes a DNA binding domain which recognizes the array. The enzymatic component of the fusion protein can then introduce damage to the array. Thus, the cell has an integrated array of DNA binding sites to which the DNA binding domain is capable of binding.

An array of DNA binding sites is any DNA region that has a binding site for a DNA binding domain, which is preferably fused to a DNA mutator enzyme domain. The array may include multiple copies of the same binding site arranged in any manner. For instance, the array may include adjacent identical or different arrays, or alternatively it may include intervening sequences between identical or different arrays. In some instance, the array specific DNA binding domain is tetR or (sc)tetR and the integrated array of DNA binding sites is a tetO array. The tetO array may be, for instance, an 85x or 240x tetO array. As shown in the Examples below, Applicants have demonstrated the utility of the claimed methods with a number of experimental variations. Some examples of combinations of components with demonstrated utility for use in the methods include:

Expression of Magl-tetR (or other DNA binding domain) with integrated 85x or 240x tetO array

Expression of single-chain (sc)tetR fused to Fokl monomer with integrated 85x or 240x tetO array, followed by 0.001%-0.008% MMS

Expression of single-chain (sc)tetR fused to scFokl dimer with integrated 85x or 240x tetO array followed by 0.001%-0.008% MMS

Expression of single-chain (sc)tetR fused to scFokl monomer with integrated 85x or 240x tetO array and overexpression of Magi

Expression of Magl-sctetR followed by MMS

Expression of Magl-sctetR followed by chloroacetaldehyde (CCA)

Expression of sctetR-Fokl followed by chloroacetaldehyde (CCA)

Once the DNA damage is created, the organism may be treated with an agent to bias the repair of the DSB to result in the creation of mutations. The bias may be achieved by targeting a pathway requiring long-range resectioning of the DSB. This results in the DNA forming a single stranded region during the biasing repair that can be selectively mutated in the single stranded region to cause targeted mutagenesis. For instance, the repair of the DSB may be biased by contacting the cell with a compound that elicits S/G2 checkpoint activation for instance by addition of chemical (i.e. MMS) or enzyme (i.e. Magi). In some instances MMS may be used, for instance, 0.001%-0.008 MMS. The methods of the invention may also be accomplished using a CRISPR(clustered regularly interspaced short palindromic repeat)/Cas9 (CRISPR-associated proteins) system. The bacterial CRISPR-Cas9 system is a promising new technology where the Cas9 nuclease is targeted by expressing a short guide RNA homologous to the target DNA sequence (Burgess DJ (2013) Technology: A CRISPR genome-editing tool. Nat Rev Genet 14:80.). The CRISPR/Cas system involves targeting of DNA with a short, complementary single- stranded RNA (CRISPR RNA or crRNA) that localizes the Cas9 nuclease to the target DNA sequence. The crRNA can bind on either strand of DNA and the Cas9 will cleave the DNA making a DSB. A trans-activating crRNA (tracrRNA) that has sequence complementary to the palindromic repeat triggers processing by the bacterial double- stranded RNA-specific ribonuclease, RNase III. crRNA and the tracrRNA can then both bind to the Cas9 nuclease, which becomes activated and specific to the DNA sequence complimentary to the crRNA.

CRISPR/Cas9 systems can be used to make the break or nick, or to deliver the MAGI to a particular region. Use of this system obviates the need for a targeting array. Following CRISPR/Cas9 manipulation, it is only required to treat cells with mutagen in order to achieve the mutagenesis. Shown in FIG. 19 is a system for converting targeted genome modification methods to targeted mutagenesis methods. The schematic in FIG. 19A shows various nuclease targeting strategies including the CRISPR/Cas9 system for introducing DSBs at specific locations within a genome.

Using the CRISPR-Cas9 system, DSBs can be generated in any sequence in the human genome that has a common RNA motif (PAM motif). A synthetic single-guide RNA (gRNA) and the Cas9 nuclease are the only components necessary to generate DSBs in cells. Efficient gene delivery of the Cas9 system can be demonstrated by monitoring GFP expression after repeated transfection and electroporation of a mixture of GFP and Cas9 plasmids. Cas9 and gRNAs can be delivered to targets within various loci in the genome of human cells. The cutting at these loci can be monitored using the SURVEYOR assay, which can detect the 1-80% efficiency. In order to initiate the mutation, the DNA-methylating agent, methyl methanesulfonate (MMS), for example, may be added. Processing of methyl groups during resectioning of DSBs will generate point mutations around DSBs. Mutation rates may be assayed using functional assays such as by loss of function of the gene. Three distinct bacterial CRISPR systems have been identified thus far, type I, II and III.

In some embodiments, the methods and products of the invention involve the use of the components of the Type II system. The Cas9 nuclease for use in eukaryotic cells may be a variation of a bacterial Cas9 nuclease that has been codon- optimized for the desired cell type. A single fused crRNA-tracrRNA construct that functions with the codon-optimized Cas9 may also be used. This single crRNA-tracrRNA fused hybrid RNA is often referred to as a guide RNA or gRNA. Within a gRNA, the crRNA portion is identified as the 'target sequence' and the tracrRNA is often referred to as the 'scaffold'. A number of online resources are available for identifying suitable target sites in desired DNA sequences.

In some embodiments the methods involve the use of a gRNA expression plasmid, which includes a form of the tracrRNA sequence (the scaffold), crRNA (target sequence) as well as a suitable promoter and necessary elements for proper processing in eukaryotic cells and a Cas9 expression plasmid, which includes the gene for Cas9 (or variation thereof) and expression elements such as a promoter. A short DNA fragment containing a target sequence may optionally be inserted into the gRNA expression plasmid. The target cells can then be doubly transfected with the gRNA expression plasmid and the Cas9 expression plasmid. Alternatively the components of the two plasmids can be combined into a single plasmid. Upon expression the single gRNA binds with and activates the Cas9 nuclease.

Genetic recombination is the breaking and rejoining of DNA strands to form new molecules of DNA encoding a novel set of genetic information. DSBs are particularly hazardous to the cell because they can lead to genome rearrangements. Three mechanisms exist to repair DSBs: non-homologous end joining (NHEJ), microhomology mediated end-joining (MMEJ), and homologous recombination. The methods of the invention involve at least in part the discovery that the repair process can be manipulated in order to promote the formation of useful mutations. In particular, the cells can be guided to promote homologous recombination and/or downregulate NHEJ, following the introduction of DSBs. One mechanism to achieve this involves the administration of a compound that elicits S/G2 (Mecl/ATR) checkpoint activation (DNA damage checkpoint).

A "compound that elicits DNA damage checkpoint activation" as used herein is a compound that causes a cell to enter S/G2, such that DNA repair processes can take place. It is also referred to as intra-S or G2 checkpoint activation or S/G2 checkpoint activation. Cell cycle checkpoints are control mechanisms used by the cell to ensure the fidelity of cell division in eukaryotic cells by confirming whether the processes at each phase of the cell cycle have been accurately completed before progression into the next phase. These compounds include for instance, chemical checkpoint activators and an enzymatic checkpoint activators. A chemical checkpoint activator is a small molecule chemical compound that induces DNA damage checkpoint activation. Chemical checkpoint activators include but are not limited to MMS, HU, nocodazole. An enzyme checkpoint activator is a protein or fragment thereof that catalyzes a reaction , specifically resulting in DNA damage checkpoint activation. Enzymatic checkpoint activators include but are not limited to Mag 1. Thus, aspects of the invention involve biasing repair through homologous recombination/resectioning by eliciting DNA damage checkpoint activation and/or genetic modifications to downregulate repair via other methods (NHEJ).

Methods to bias the repair of the DSB to result in the creation of mutations may alternatively or additionally be achieved by downregulating non-HR processes, (such as NHEJ and de novo telomere addition) in the cell. Methods for downregulating non-HR processes include for instance, genetic modification, inhibitory nucleic acids, and chemicals. Chemicals include NDA-PKcs inhibitors that selectively downregulate NHEJ such as NU7026 and Vanillin as well as wortmannin. Genetic modifications and inhibitory nucleic acids useful in these methods are those which target a critical component in the non-HR pathways.

The choice between NHEJ and homologous recombination for repair of a double- strand break is regulated at the initial step in recombination, 5' end resection. In this step, the 5' strand of the break is degraded by nucleases to create long 3' single- stranded tails. DSBs that have not been resected can be rejoined by NHEJ, but resection of even a few nucleotides strongly inhibits NHEJ. Nonhomologous recombination or NHEJ involves DNA Ligase IV, a specialized DNA ligase that forms a complex with the cof actor XRCC4, and directly joins the two ends of a DSB using short homologous sequences called microhomologies present on the single- stranded tails of the DNA ends to be joined. NHEJ in eukaryotes involves a number of proteins involved in various steps, each of which can be targeted as part of the down regulation or genetic manipulation of the NHEJ process. For instance, NHEJ involves End binding and tethering, End processing, Ligation, and Regulation.

The step of end binding and tethering involves the Mrel l-Rad50-Xrs2 (MRX) complex in yeast or the corresponding mammalian complex of Mrel 1-Rad50-Nbsl (MRN). DNA-PKcs is also thought to participate in end bridging during mammalian NHEJ. Eukaryotic Ku is a heterodimer consisting of Ku70 and Ku80, and forms a complex with DNA-PKcs, which is present in mammals but absent in yeast. Ku is known to interact with the DNA ligase IV complex and XLF. Mice having knocked out Ku or DNA-PKcs have been developed and are viable.

End processing involves removal of damaged or mismatched nucleotides by nucleases and resynthesis by DNA polymerases. The X family DNA polymerases Pol λ and Pol μ (Pol4 in yeast) fill gaps during this process.

The ligation step involves DNA ligase IV complex (catalytic subunit DNA ligase IV and its cofactor XRCC4 (Dnl4 and Lifl in yeast)). XLF, also known as Cernunnos, also plays a role in this process.

The regulation step is mediated by cyclin-dependent kinase Cdkl (Cdc28 in yeast).

In certain embodiments inhibitors of non-HR processes are inhibitory nucleic acids such as short interfering nucleic acid or antisense oligonucleotides specific for a gene transcript of a protein involved in a non-HR process. The inhibitory nucleic acids reduce the amount of mRNA specific for the non-HR protein in a cell of interest.

Inhibitor molecules that are short interfering nucleic acids (siNA), which include, short interfering RNA (siRNA), double- stranded RNA (dsRNA), micro-RNA (miRNA), and short hairpin RNA (shRNA) molecules are used to inhibit the expression of target genes. The siNAs of the present invention, for example siRNAs, typically regulate gene expression via target RNA transcript cleavage/degradation or translational repression of the target messenger RNA

(mRNA). In one embodiment siRNAs are exogenously delivered to a cell. In a specific embodiment siRNA molecules are generated that specifically target a non-HR protein. A short interfering nucleic acid (siNA) of the invention can be unmodified or chemically-modified. A siNA of the instant invention can be chemically synthesized, expressed from a vector or enzymatically synthesized. The instant invention also features various chemically-modified synthetic short interfering nucleic acid (siNA) molecules capable of inhibiting gene expression or activity in cells by RNA interference (RNAi). The use of chemically-modified siNA improves various properties of native siNA molecules through, for example, increased resistance to nuclease degradation in vivo and/or through improved cellular uptake. Furthermore, siNA having multiple chemical modifications may retain its RNAi activity. For example, in some cases, siRNAs are modified to alter potency, target affinity, the safety profile and/or the stability to render them resistant or partially resistant to intracellular degradation. Modifications, such as phosphorothioates, for example, can be made to siRNAs to increase resistance to nuclease degradation, binding affinity and/or uptake. In addition, hydrophobization and bioconjugation enhances siRNA delivery and targeting (De Paula et al., RNA. 13(4):431-56, 2007) and siRNAs with ribo-difluorotoluyl nucleotides maintain gene silencing activity (Xia et al., ASC Chem. Biol. l(3):176-83, (2006). siRNAs with amide-linked oligoribonucleosides have been generated that are more resistant to SI nuclease degradation (Iwase R et al. 2006 Nucleic Acids Symp Ser 50: 175-176). In addition, modification of siRNA at the 2'-sugar position and phosphodiester linkage confers improved serum stability without loss of efficacy (Choung et al., Biochem. Biophys. Res. Commun. 342(3):919-26, 2006). In one study, 2'-deoxy-2'-fluoro-beta-D-arabinonucleic acid (FANA)-containing antisense

oligonucleotides compared favorably to phosphorothioate oligonucleotides, 2'-0-methyl- RNA/DNA chimeric oligonucleotides and siRNAs in terms of suppression potency and resistance to degradation (Ferrari N et a. 2006 Ann N Y Acad Sci 1082: 91-102).

In some embodiments an siNA is an shRNA molecule encoded by and expressed from a genomically integrated transgene or a plasmid-based expression vector. Thus, in some embodiments a molecule capable of inhibiting gene expression is a transgene or plasmid-based expression vector that encodes a small-interfering nucleic acid. Such transgenes and expression vectors can employ either polymerase II or polymerase III promoters to drive expression of these shRNAs and result in functional siRNAs in cells. The former polymerase permits the use of classic protein expression strategies, including inducible and tissue-specific expression systems. In some embodiments, transgenes and expression vectors are controlled by tissue specific promoters. In other embodiments transgenes and expression vectors are controlled by inducible promoters, such as tetracycline inducible expression systems. Examples of making and using such hairpin RNAs for gene silencing in mammalian cells are described in, for example, (Paddison et al., Genes Dev, 2002, 16:948-58; McCaffrey et al., Nature, 2002, 418:38-9;

McManus et al., RNA 2002, 8:842-50; Yu et al, Proc Natl Acad Sci USA, 2002, 99:6047-52).

Non-HR proteins which may be targeted for down regulation include but are not limited to: DNA-PKcs, Ku70, u80, a protein of DNA ligase IV complex, XRCC4, Dnl4, Lif 1, XLF, Cdkl, and Cdc28.

Increased mutation rates is achieved by induction of mutagenesis according to the invention in arrested cells. Knowledge of the underlying mechanism behind mutations created by these methods allows for new ways of using Magl-sctetR, sctetR-Fokl, and other targeted mutators. Targeted mutagenesis relies on homologous recombination, long range resectioning, and base pair damage. This invention includes methods to increase targeted mutation rates by altering the cellular context in such a way that these processes are favored. In addition to increased base pair damage by chemical agents like MMS and CAA, targeted mutagenesis can be increased by altering the growth and or mutator expression timing of the population being mutated.

One way to do this is to grow up a large population of cells without expressing the mutator and then arrest them all at the G2/M cell cycle checkpoint while turning on expression of the mutator. HR proteins are upregulated in these post-replication cells, and, because the cell cycle is stopped, there is sufficient time for long range resectioning to occur. More cells on average experience a clustered damage event that leads to HR because there is more time for clustered damage to occur. Ways to arrest at the G2/M DNA damage checkpoint include DNA damage checkpoint chemicals such as nocodazole and mutants in cell cycle kinases like CDC 15 or CHK1 in yeast. This invention also encompasses arrest at the intra-S checkpoint used in the manner described above with chemicals like hydroxyurea or camptothecin. For example a DNA mutator may be overexpressed specifically in the S/G2 phases of the cell cycle using a cell-cycle dependent promoter, such as HHOl and HH02 (histone genes in yeast).

Another way to alter the cellular context is to grow the cells in concentrations of the aforementioned cell cycle arresting chemicals that do not lead to full cell cycle arrest but instead cause only an extension of the S and/or G2 phases of the cell cycle. In this scenario the mutator construct is expressed continuously, and the mutation rate is increased because cells spend longer in a cell cycle phase were HR is the dominant repair pathway. Also, a clustered damage event and long range resectioning are more likely to occur because of slow cell cycle progression. Non-HR protein -specific siRNAs and shRNAs are commercially available. The present invention, thus, contemplates in vitro use of siRNAs (shRNAs, etc.) as well as in vivo pharmaceutical preparations containing siRNAs (shRNAs, etc.) that may be modified siRNAs (shRNAs, etc.) to increase their stability and/or cellular uptake under physiological conditions, that specifically target nucleic acids encoding proteins involved in the non-HR pathway, together with pharmaceutically acceptable carriers.

In certain embodiments inhibitors of non-HR processes are antisense nucleic acids. Antisense nucleic acids include short oligonucleotides as well as longer nucleic acids.

Preferably the antisense nucleic acids are complementary to and bind to portions of the proteins involved in non-HR processes coding sequence or 5' nontranslated sequence, thereby inhibiting translation of functional polypeptides. Thus the invention embraces antisense oligonucleotides that selectively bind to a nucleic acid molecule encoding proteins involved in non-HR processes, to reduce the expression (transcription or translation) of these proteins. As used herein, the term "antisense oligonucleotide" describes an oligonucleotide that is an oligoribonucleotide, oligodeoxyribonucleotide, modified oligoribonucleotide, or modified oligodeoxyribonucleotide which hybridizes under physiological conditions to DNA and, thereby, inhibits the transcription of that gene and/or the translation of that mRNA.

Those skilled in the art will recognize that the exact length of the antisense

oligonucleotide and its degree of complementarity with its target will depend upon the specific target selected, including the sequence of the target and the particular bases which comprise that sequence. It is preferred that the antisense oligonucleotide be constructed and arranged so as to bind selectively with the target under physiological conditions, i.e., to hybridize substantially more to the target sequence than to any other sequence in the target cell under physiological conditions. Based upon the sequences of nucleic acids, including allelic or homologous genomic and/or cDNA sequences, one of skill in the art can easily choose and synthesize any of a number of appropriate antisense molecules for use in accordance with the present invention. In order to be sufficiently selective and potent for inhibition, such antisense oligonucleotides should comprise at least 10 and, more preferably, at least 15 consecutive bases which are complementary to the target, although in certain cases modified oligonucleotides as short as 7 bases in length have been used successfully as antisense oligonucleotides (Wagner et al., Nature Biotechnol. 14:840-844, 1996). Most preferably, the antisense oligonucleotides comprise a complementary sequence of 20-30 bases.

In another embodiment, the antisense nucleic acids of the invention may be produced by expression in cells by expression vectors introduced therein. The choice and design of an appropriate vector is within the ability and discretion of one of ordinary skill in the art.

Once the single stranded region is formed the cell can be treated to promote mutation. Any compound that produces mutations in single stranded DNA can be used to achieve this step. For instance, chemicals that elicit single stranded DNA (ssDNA) damage may be used to promote the mutation. Examples of chemicals that elicit ssDNA damage include but are not limited to MMS, chloroacetaldehyde, sodium bisulfate and osmium tetraoxide. Enzymes that perform this function include but are not limited to APOBEC family of RNA editing enzymes, AID. As used herein "activation-induced cytidine deaminase" or ("AID") refers to members of the AID/ APOBEC family of RNA/DNA editing cytidine deaminases capable of mediating the deamination of cytosine to uracil within a DNA sequence. (U.S. Pat. No. 6,815,194). The term "AID homolog" refers to the enzymes of the Apobec family and include, for example, Apobec and, in particular, can be selected from Apobec family members such as Apobec- 1 , Apobec3C or Apobec3G.

The methods are performed in vivo in an organism or cell. The organism or cell may be any organism or cell in which a DNA binding array and adjacent genes to be mutated can be introduced. For example, organisms and cells according to the invention include prokaryotes and eukaryotes (i.e. yeast, plants). Prokaryotes include but are not limited to Cyanobacteria, Bacillus subtilis, E, coli, Clostridium, and Rhodococcus. Eukaryotes include, for instance, algae (N anno chlorop sis), yeast such as, S. cerevisiae and P. pastoris, mammalian cells, such as for instance human cells, including HEK 293 cells, primary neuronal stem cell lineages, embryonic stem cells, adult stem cells, and rodents, plants including for instance Arabidopsis. In some embodiments, the cell is not a cell of human B lymphocyte lineage. The methods of the invention can be achieved in some instance using fusion proteins or recombinant nucleic acids designed to produce fusion proteins. For instance the fusion proteins maybe composed of a DNA binding domain fused to a DNA mutator enzyme domain. Some examples of fusion proteins useful according to the invention include: (sc)tetR-Fokl, TAL-Fokl, and Magl-sctetR.

The fusion proteins may be prepared using recombinant DNA technology. For instance, the genes for the two or more domains are assembled and inserted into a vector such as a plasmid. The vectors are then used to transfect the target cell where the gene products are expressed as a fusion protein and can enter the nucleus to access the genome. Thus, the invention also encompasses, fusion proteins, isolated nucleic acid sequences that encode any of the fusion proteins, expression vectors comprising the isolated nucleic acids, and host cells comprising the expression vectors, isolated nucleic acids and or fusion proteins.

The methods of the invention may be used to probe evolving protein-protein interactions. Protein-protein interactions are inherently multigene because they involve a binding surface that is composed of multiple proteins. One of the most well studied protein-protein interactions is that of the leucine zipper family of transcription factors exemplified by GCN4 in S. cerevisiae (Kerppola and Curran, 1991). Basic leucine zipper domains (bZIPs) are conserved 50-65 amino acid regions that form an a-helical dimerization interface that also interacts with DNA

(Ellenberger et al., 1992; O'Shea et al., 1991). bZIP domains exist in both homo- and hetero- dimeric form in S. cerevisiae, and their complete interaction map has been determined

(Deppmann et al., 2006). Within this map there are bZIP pairs which exhibit both weak and strong interactions, making the evolution of stronger interaction between a weak pair both achievable and easier to detect. Specifically, mutagenesis of the GCN4 homodimer has led to altered binding strength (Hu et al., 1993), demonstrating that this type of evolution can be successful.

One benefit of studying protein-protein interactions in the methods of the invention is that they are easily selected for using a synthetic yeast two hybrid (Y2H) selection scheme.

Fusion of one bZIP partner to the DNA binding domain of the bacterial repressor lexA and the other bZIP partner to the VP 16 acidic activation domain (Fig. 20) will tie the strength of interactions between the two bZIP domains to the level of gene expression. Promoters using 4 tandem copies of the lex operator from the ColEl origin upstream of the minimal CYCl TATA box have been shown to robustly drive expression in Y2H systems. The methods can be achieved using lexO based promoters to drive multiple selection markers, making off-target mutations that allow cells to evade selection even less likely. HIS3 is used as a selection marker because it can be inhibited by the small molecule 3-aminotriazole (3- AT), allowing for a linear relationship between growth in media lacking histidine and HIS3 expression for a wide range of both growth and expression levels. The second marker used is YFP, which allows for fluorescence activated cell sorting (FACS) of only the brightest cells. FACS represents a convenient way to discard the majority of the culture in which there is no improvement, avoiding the bottleneck created by repeated growth in batch culture.

bZIP fusions are under methinonine control from the MET3pr and MET14pr, which express at similar levels (Korch et al., 1991) in media with decreased levels of methionine. The graded expression of bZIP constructs is used to ensure that selection occurs at a protein concentration where reporter expression is dominated by the affinity of the bZIP interaction. The cultures that perform best at the highest methionine concentration (lowest expression level) are assumed to have the bZIP domains with greatest affinity. In order to monitor expression of each bZIP construct during evolution, they are fused with either CFP or mCherry. This will also allow us to confirm the methionine responsiveness of each promoter in a construct dependent manner.

Evolution is carried out in sequential steps of mutation and selection. Mutagenesis will be carried out in an arrested population of 108 cells with the addition of an ssDNA specific damaging agent like CAA. After arrest, cells are recovered for ~3 generations in fresh media to decrease any mutant growth defects incurred by the mutation process. The expanded population is then diluted into media lacking histidine to select for greater HIS3 expression. 24 populations containing various levels of 3-AT and methionine will insure that cells undergo the most stringent selection at a bZIP expression level where affinity dominates the level of HIS3 expression. Each population is then sorted using FACS and the brightest 1% of cells from each population are retained. The 1% of cells at the highest methionine level that is the brightest as compared to a control containing dox are chosen to repeat the mutagenesis process. A control lacking Magl-sctetR will allow for the detection of spontaneous mutations that confer increased fitness.

A detailed process to implement this sequential mutagenesis and selection for example includes the following steps. 108 cells are arrested at an OD of 0.5 in 20 mL using nocodazole and CAA for 4 hours. Cells are then be washed and resuspended in 24 mL of fresh media and allowed to grow to saturation (109) cells. 1 mL aliquots are added to 9 mL of selection media in a deep 24- well plate. Columns will include variable levels of 3-AT and rows include variable levels of methionine. Cultures will be allowed to grow to saturation, and then 1% of cells in each culture are retained during sorting by FACS. The 1% of cells with the highest average YFP expression at the highest methionine level in comparison to the control are then regrown to an OD of 0.5 in 20 mL and the process repeated. Periodically, selected cultures are assessed for the continuing function of the TaGTEAM system by measuring the mutation rate at the nearby ade2-l marker.

As used herein the term "isolated nucleic acid molecule" refers to a nucleic acid that is not in its natural environment, for example a nucleic acid that has been (i) extracted and/or purified from a cell, for example, an algae, yeast, plant or mammalian cell by methods known in the art, for example, by alkaline lysis of the host cell and subsequent purification of the nucleic acid, for example, by a silica adsorption procedure; (ii) amplified in vitro, for example, by polymerase chain reaction (PCR); (iii) recombinantly produced by cloning, for example, a nucleic acid cloned into an expression vector; (iv) fragmented and size separated, for example, by enzymatic digest in vitro or by shearing and subsequent gel separation; or (v) synthesized by, for example, chemical synthesis. In some embodiments, the term "isolated nucleic acid molecule" refers to (vi) an nucleic acid that is chemically markedly different from any naturally occurring nucleic acid. In some embodiments, an isolated nucleic acid can readily be manipulated by recombinant DNA techniques well known in the art. Accordingly, a nucleic acid cloned into a vector, or a nucleic acid delivered to a host cell and integrated into the host genome is considered isolated but a nucleic acid in its native state in its natural host, for example, in the genome of the host, is not. An isolated nucleic acid may be substantially purified, but need not be. For example, a nucleic acid that is isolated within a cloning or expression vector is not pure in that it may comprise only a small percentage of the material in the cell in which it resides. Such a nucleic acid is isolated, however, as the term is used herein.

Some aspects of this invention relate to nucleic acids, encoding a gene product of a fusion protein, which are linked to a promoter or other transcription activating element. In some embodiments, the nucleic acid encoding the gene product and linked to a promoter is comprised in an expression vector or expression construct. As used herein, the terms "expression vector" or "expression construct" refer to a nucleic acid construct, generated recombinantly or

synthetically, with a series of specified nucleic acid elements that permit transcription of a particular nucleic acid in a host cell, for example, S. cerevisiae. In some embodiments, the expression vector may be part of a plasmid, virus, or nucleic acid fragment. In some embodiments, the expression vector includes the coding nucleic acid to be transcribed operably linked to a promoter. A promoter is a nucleic acid element that facilitates transcription of a nucleic acid to be transcribed. A promoter is typically located on the same strand and upstream (or 5' ) of the nucleic acid sequence the transcription of which it controls. In some embodiments, the expression vector includes the coding nucleic acid to be transcribed operably linked to a heterologous promoter. A heterologous promoter is a promoter not naturally operably linked to a given nucleic acid sequence.

In some embodiments, the expression vector includes a coding nucleic acid, for example, a nucleic acid encoding a fusion protein described herein, operably linked to a constitutive promoter. The term "constitutive promoter" refers to a promoter that allows for continual transcription of its associated gene. In some embodiments, the expression vector includes a nucleic acid operably linked to an inducible promoter. The term "inducible promoter", interchangeably used herein with the term "conditional promoter", refers to a promoter that allows for transcription of its associated gene only in the presence or absence of biotic or abiotic factors. Drug-inducible promoters, for example tetracycline/doxycycline inducible promoters, tamoxifen-inducible promoters, as well as promoters that depend on a recombination event in order to be active, for example the cre-mediated recombination of loxP sites, are examples of inducible promoters that are well known in the art.

Methods to deliver expression vectors or expression constructs into cells, for example, into yeast cells, are well known to those of skill in the art. Nucleic acids, including expression vectors, can be delivered to prokaryotic and eukaryotic cells by various methods well known to those of skill in the relevant biological arts. Methods for the delivery of nucleic acids to a cell in accordance to some aspects of this invention, include, but are not limited to, different chemical, electrochemical and biological approaches, for example, heat shock transformation,

electroporation, transfection, for example liposome-mediated transfection, DEAE-Dextran- mediated transfection or calcium phosphate transfection. In some embodiments, a nucleic acid construct, for example an expression construct comprising a fusion protein nucleic acid sequence, is introduced into the host cell using a vehicle, or vector, for transferring genetic material. Vectors for transferring genetic material to cells are well known to those of skill in the art and include, for example, plasmids, artificial chromosomes, and viral vectors. Methods for the construction of nucleic acid constructs, including expression constructs comprising constitutive or inducible heterologous promoters, knockout and knockdown constructs, as well as methods and vectors for the delivery of a nucleic acid or nucleic acid construct to a cell are well known to those of skill in the art, and are described, for example, in J. Sambrook and D. Russell, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press; 3rd edition (January 15, 2001); David C. Amberg, Daniel J. Burke; and Jeffrey N. Strathern, Methods in Yeast Genetics: A Cold Spring Harbor Laboratory Course Manual, Cold Spring Harbor Laboratory Press (April 2005); John N. Abelson, Melvin I. Simon, Christine Guthrie, and Gerald R. Fink, Guide to Yeast Genetics and Molecular Biology, Part A, Volume 194 (Methods in Enzymology Series, 194), Academic Press (March 11, 2004); Christine Guthrie and Gerald R. Fink, Guide to Yeast Genetics and Molecular and Cell Biology, Part B, Volume 350 (Methods in Enzymology, Vol 350), Academic Press; 1st edition (July 2, 2002); Christine

Guthrie and Gerald R. Fink, Guide to Yeast Genetics and Molecular and Cell Biology, Part C, Volume 351, Academic Press; 1st edition (July 9, 2002); Gregory N. Stephanopoulos, Aristos A. Aristidou and Jens Nielsen, Metabolic Engineering: Principles and Methodologies, Academic Press; 1 edition (October 16, 1998); and Christina Smolke, The Metabolic Pathway Engineering Handbook: Fundamentals, CRC Press; 1 edition (July 28, 2009), all of which are incorporated by reference herein.

Some aspects of this invention relate to engineering of a cell, for example, S. Cerevisiae, to express fusion proteins and/or to disrupt non-HR processes to enhance DNA mutations. Thus in some embodiments, the combination of genetic modifications includes a push modification and a pull modification. In some embodiments, the push modification comprises delivery of a fusion protein to promote DSBs. In some embodiments, the pull modification is a genetic modification that decreases the level of a product involved in the non-HR processes of the cell.

The invention also includes in some aspects, plants having a mutated germline. The plants can be produced according to the methods described herein. As discussed above a set of methods have been developed according to the invention to target mutagenesis to a particular genomic region. Targeting is accomplished by first selectively introducing a double- strand break (DSB) in the region of interest and then biasing repair of the DSB through a pathway requiring long-range resectioning of the broken DNA. The ssDNA which is exposed during resectioning is then mutated. The methods can involve novel products and their development and use can be enhanced through the use of kits. Accordingly, the invention also relates to kits and products for accomplishing the methods of the invention. The present invention is further illustrated by the following Examples, which in no way should be construed as further limiting. The entire contents of all of the references

(including literature references, issued patents, published patent applications, and co pending patent applications) cited throughout this application are hereby expressly incorporated by reference.

EXAMPLES

Materials and Methods:

Plasmid and yeast strain construction. Plasmids and yeast strains used in this study are listed in Table 4 and 5. A complete primer list is given in Table 6 (SEQ ID NO: 2 through SEQ ID NO: 108).

Table 4: Strain List

Integrating plasmid/

Name Parent Genetic change PCR primers Usage notes

MATa ade2-l trpl-1 W303 base strain, confirmed to be RAD5 using the protocol canl-100 Ieu2-3,112 his recommended by the SGD community wiki

NY0003 N/A 3-11,5 ur 3 GAL+ N/A (http://wiki.yeastgenome.org/index.php/CommunityW303.html).

NY0339 NY0003 canl-100::KIURA3 primers 1 and 2

NY0343 NY0339 canl-100::CANl primers 3 and 4

NY0378 NY0343 APNl-.-.KanR primers 5 and 6

Integration of

NY0389 NY0343 CHRI197000::pNB0537 pNB0537

Integration of

NY0390 NY0378 CHRI197000::pNB0537 pNB0537

NY0526 NY0389 CHRI180000::K7Ufi,43 primers 7 and 8 centromeric side distance dependence

NY0544 NY0389 CHRI189000::K/Ufl,43 primers 9 and 10 centromeric side distance dependence

NY0545 NY0389 CHRI192000::K7Ufi,43 primers 11 and 12 centromeric side distance dependence

NY0554 NY0389 pNB0537::MUR43 primers 13 and 14 APN1 targeted mutagenesis test strain

NY0542 NY0389 CHRI209000::K7Ufi,43 primers 15 and 16 telomeric side distance dependence

NY0543 NY0389 CHRI213000::K7Ufl,43 primers 17 and 18 telomeric side distance dependence

NY0520 NY0343 CHRI180000::K7Ufi/13 primers 7 and 8 distance dependence in WT

NY0519 NY0343 CHRI197000::K7Ufi,43 primers 19 and 20 distance dependence in WT

NY0541 NY0343 CHRI213000::/ /Ufl ¾3 primers 17 and 18 distance dependence in WT

NY0612 NY0339 ade2-l::CgJRPl primers 21 and 22 clean delete of entire ade2 cassette

Integration of

NY0619 NY0624 his3-l J,5::pNB0603 pNB0603 plasmid targeted mutagenesis test strain

NY0620 NY0624 /?/^'s3-22,5::pRS303 Integration of pRS303 plasmid targeted mutagenesis test strain control

NY0737 NY0544 ETOl::KanR Primers 23 and 24

NY0739 NY0554 ETOiiiKanR Primers 23 and 24

Integration of

NYQ763 NY0343 CHRI197000::pNB0673 PNB0673

NY0775 NY0544 SGS1 ::CgTRPl Primers 25 and 26

NY0777 NY0554 SCSI ::CgTRPl Primers 25 and 26

NY0873 NY0763 pNB0673::WUft43 primers 13 and 14 Ox array test strain

NY0874 NY0389 CHRI118000::K7Ufi,43 Primers 27 and 28 centromeric side distance dependence

RAD52::RAD52-CFP-

NY0883 NY0554 KanR Primers 29 and 30 240x array Rad52-CFP strain

RAD52::RAD52-CFP-

NYQ885 NY0873 KanR Primers 29 and 30 no array Rad52-CFP strain Integrating plasmid/

Name Parent Genetic change PCR primers Usage notes

NY0894 NY0873 R£V3;.CgTRPl Primers 31 and 32

NYQ895 NY0873 RAD51: CgTRPl Primers 33 and 34

NY0896 NY0873 RAD52: CgTRPl Primers 35 and 36

NY0897 NY0873 RAD59:;CgTRPl Primers 37 and 38

NY0898 NY0873 POL32::CgTRPl Primers 39 and 40

NY0899 NY0873 fi£22::CgTRPl Primers 41 and 42

NY0900 NY0873 K USO.-.CgTRPl Primers 43 and 44

NY0901 NY0544 R£V3::CgTRPl Primers 31 and 32

NYQ901 NY0544 RAD51: CgTRPl Primers 33 and 34

NY0903 NY0544 RAD52: CgTRPl Primers 35 and 36

NY0904 NY0544 RAD59:;CgTRPl Primers 37 and 38

NY0906 NY0544 POL32::CgTRPl Primers 39 and 40

NY0907 NY0544 fi£22::CgTRPl Primers 41 and 42

NY0908 NY0544 K USO.-.CgTRPl Primers 43 and 44

NY0909 NY0554 R£V3::CgTRPl Primers 31 and 32

NYQ910 NY0554 RAD51: CgTRPl Primers 33 and 34

NY0911 NY0554 RAD52: CgTRPl Primers 35 and 36

NY0912 NY0554 RAD59:;CgTRPl Primers 37 and 38

NY0913 NY0554 POL32::CgTRPl Primers 39 and 40

NY0914 NY0554 fi£22::CgTRPl Primers 41 and 42

NY0915 NY0554 v/cusa-.-cgTRPi Primers 43 and 44

NY0923 NY0873 F 02::KanR Primers 23 and 24

NYQ924 NY0873 SGS1 ::CgTRPl Primers 25 and 26

Integration of

NY0927 NY0343 CHRI197000::pNB0775 pNB0775 85x no homology test strain

NY0931 NY0737 SCSI ::CgTRPl Primers 25 and 26

NY0932 NY0739 SCSI ::CgTRPl Primers 25 and 26

NY0951 NY0554 SMU.vCgTRPl Primers 45 and 46

NY0971 NY0951 DDC2;;KanR Primers 47 and 48

- Yeast transformations were performed using the method in (Daniel Gietz and Woods)

- All distances on chromosome 1 correspond to positions in the reference sequence (S288C background). W303 differs significantly in this region from the reference sequence, and primers were designed using the known W303 sequence (Liti et al., 2009)

(http://www.sanger.ac.uk/research/projects/genomeinformatics/sgrp.html). Distances were confirmed by PCR (primers 48, 54, and 100-103) from one position to the next.

- Clean delete means deletion of the promoter, ORF, and terminator of a gene so as to remove any possible homology for marker recombination during fluctuation analysis.

Table 5:

Plasmid List - 34 -

Cloning Insert PCR Addgene

Name Method Backbone Insert(s) primers deposited Usage notes

Plasmids used

pLAU44 Kind gift of D. Sherrat (Lau et al., 2003)

pRS4Dl Kind gift of J. Collins (Blake et al., 2003)

pYES-MAG Kind gift of L. Samson (Glassner et al., 1998)

pWH610(B+sB) Kind gift of W. Hillen (Krueger et al., 2003)

Plasmids constructed and used

GALlpr

pNB0298 Ligation PRS415 (XhoI/BamHI) (XhoI BamHI) 81 and 82 no pGALlpr pNB0437 Ligation pNB0298 (Spel/Sacl) MAGI (Spel/Sacl) 83 and 84 no

pNB0443 Ligation pNB0437 (Sall/SacI) AC77i(SalI/SacI) 85 and 86 no

pNB0443

pNB0450 Ligation (NgoMIV/XhoI) none (blunted) N/A no MAGI sctetR binding test by fluorescence pNB0451 Ligation pRS4Dl (Notl/Sacl) none (blunted) N/A no knockdown pNB0470 Gap repair pNB0450 (Spel/Sall) sctetR 88 and 87 no sctetR pNB0471 Gap repair pNB0450 (Sall/Notl) vYFP 89 and 90 no MAGl-v FP plasmid pNB0472 Gap repair pNB0450 (Sall/Notl) sctetR 91 and 87 yes /VMGl-sctetR

PNB0476 Gap repair pNB0450 (Spel/Sall) vYFP 92 and 90 no NLS-vYFP

MAG1- pNB0602 Gap repair pNB0450 (Sall/Notl) sctetR-cYFP 91 and 90 no sctetR-cYFP pNB0298 (Xhol/Spel)

and pNB0472 integrated pNB0603 Ligation PRS303 (Xhol/Sacl) (Spel/Sacl) N/A yes /VMGI-sctetR

CHRI 5' homology integrated (Ascl/Xbal) and CHRI 3' 240x tetO pNB0537 Ligation pLAU44 (Notl/Xbal) homology (Notl/Ascl) 93 to 96 yes array pNB0568 Ligation pBS (Notl/Xbal) pNB0537 (Notl/Xbal) N/A no

240x teto array plasmid 240x pNB0586 Ligation pRS316 (Xbal/Xhol) (Xbal/Xhol) N/A yes tetO array plasmid 240x tetO array pNB0640 Ligation pNB0586 (Xhol) ade2-l cassette (Xhol) 97 and 98 no w/ade2-l pNB0653 Ligation pBS (Apal/Hindlll) KIURA3 cassette 99 and 100 no

Fokl (Bibikova et al.,

pNB0663 Ligation pNB0450 (BamHI/Sall) 2003) (BamHI/Xhol) N/A no

PNB0665 Gap repair ΡΝΒ0663 (BamHI) sctetR 88 and 103 yes sctetR-FOKI integrated Ox pNB0673 Ligation pNB0537 (Xhol/Xbal) none (blunted) N/A no tetO array pNB0763 Ligation pBS (EcoRI/Xmal) pNB0537 (EcoRI/Xmal)

pNB0773 Ligation pNB0763 (Notl/Xbal) pNB0568 (Notl/Xbal) N/A no integrated

85x array pNB0773 pNB0653 w/o pNB0775 Ligation (NgoMIV/Hindlll) (NgoMIV/Hindlll) N/A yes homology pNB0298 (Xhol/Agel)

and pNB0665

pNB0784 Ligation pNB0298 (Xhol/Sacl) (Agel/Sacl) N/A no

coexpression of sctetR- FOKI and pNB0785 Gap repair pNB0784 (Xhol) MAGl-ACTlt 101 and 102 no MAGI pNB0841 Ligation p S306 (Ndel/Ncol) fragment of URA3 104 and 105 no

integration of TDH3pr- Magl-sctetR

7DH3pr-Magl-sctetR- to delete

PNB0843 Ligation PNB0841 (Nhel/Ascl) ACTlt 106 and 107 yes URA3

Table 6: Primer List (SEQ ID NO: 2 through SEQ ID NO: 108)

SE

Q ID

N

Name Sequence O: Template

Integrating

primers

CgCanlKO( tcttcagacttcttaactcctgtaaaaacaaaaaaaaaaaaaggcatagc CACAGGAAAC 2

1 +) AGCTATGACC KIURA3 on pBluescript

CgCanlKO( agaatgcgaaatggcgtggaaatgtgatcaaaggtaataaaacgtcatat GTTGTAAAAC 3

2 -) GACGGCCAGT

CANlinsv2( 4

3 +) GGTTGCGAACAGAGTAAACCGAATCAGGG CAN1

CANlinsv2( 5

4 -) GCTTCTACTCCGTCTGCTTTCTTTTCGGG

APN1KO- ATGCCTTCGACACCTAGCTTTGTTAGATCTGCTGTCTCGAAATACAAAT 6

5 Kanv2(+) T GATCTGTTTAGCTTGCCTCGTCCC pNB0132

APN1KO- TTATTCTTTCTTAGTCTTCCTCTTCTTTGTCATTTGTGACAAGATATCAT 7

6 Kanv2(-) AAACTGGATGGCGGCGGTTAG

URA- GTTAGTTAGTTACTGTTAGGACGCTTCGGCGAGCTGATGTCTGACTTCT 8

7 17kb(+) C CACTATAGGGCGAATTGGGTAC KIURA3 on pBluescript

URA-17kb(- TTACGGCCATTATCAGCGGTAAAACACCCAAGGTGTTGACTAAGTGAT 9

8 ) GG AAAGGGAACAAAAGCTGGAGC

agatttccaagcaagcttttagtggaaatcatcgcgcgcaagccagcggt 10

9 URA-8kb(+) CACTATAGGGCGAATTGGGTAC KIURA3 on pBluescript

1 TCCGCACGTCCTACGTTTAGAAAGTAACGATGCCAATCTTCATCACGGT 11

0 URA-8kb(-) A AAAGGGAACAAAAGCTGGAGC

1 TTTGGAAGTGACTGGCGCCGCCGCTGGCTACTATAATAGCAGCGACTG 12

1 URA-5kb(+) TA CACTATAGGGCGAATTGGGTAC KIURA3 on pBluescript

1 TTGGTGCACGTTCGCTCGGCGAGTAAAAGAGGTAATCCAAACGACGGG 13

2 URA-5kb(-) AT AAAGGGAACAAAAGCTGGAGC

1 actgcccgctttccagtcgggaaacctgtcgtgccagctgcattaatgaa 14

3 URAsfm(+) CACTATAGGGCGAATTGGGTAC KIURA3 on pBluescript

1 URAsfm(-) GCACACAGCCCAGCTTGGAGCGAACGACCTACACCGAACTGAGATACC 15 TA AAAGGGAACAAAAGCTGGAGC

U A3kbv3( atcgaaataaaatgctgtatcacgggcgattattccatggcgaaatgagg 16

+) CACTATAGGGCGAATTGGGTAC KIURA3 on pBluescript

URA3kbv4( GGTGTTAGATACGGATGTGAAAGGGCGATAAGACATTTGGAAGTTAAT 17

-) GA AAAGGGAACAAAAGCTGGAGC

URAllkbv2 gcagtctttacacttctggcactaattaatgtggcctcaggagccacaga 18

(+) CACTATAGGGCGAATTGGGTAC KIURA3 on pBluescript

GAATACTGGTAAAAATTTATATTCATCCCACTTTTCCTCTGGCCTGCTG 19

URAllkb(-) G AAAGGGAACAAAAGCTGGAGC

URA0kbv4( tgcaaaaattttagccgcaaatctcatcttcagtcccacatgattcacctg 20

+) CACTATAGGGCGAATTGGGTAC KIURA3 on pBluescript

URA0kbv4( AATAGTTGTGACGTTTAGCAGCTCGACGCTGCATGAAAATCTGCAGAA 21

-) AA AAAGGGAACAAAAGCTGGAGC

CgKO- gcgcactaccagtatatcatctcatttccgtaaataccaaatgtattata CACAGGAAAC 22

ADE2(+) AGCTATGACC CgTRPl on pBluescript

CgKO- ATTGAGCCGCCTTATATGAACTGTATCGAAACGTTATTTTTTTAATCGC 23

ADE2(-) A GTTGTAAAAC GACGGCCAGT

PrKO- ATGGGTATCC AAGGTCTTCT TCCTCAGTTA AAGCCCATAC 24

EX01(+) AGAATCCAGT GATCTGTTTAGCTTGCCTCGTCCC pNB0132

PrKO- TTTATAAACAAATTGGGAAAGCAAGGAGATAGATCTGACTGCCGGCCG 25

EXOl(-) AG AAACTGGATGGCGGCGGTTAG

CgKO- ATGGTGACGA AGCCGTCACA TAACTTAAGA AGGGAGCACA 26

SGS1(+) AATGGTTAAA CACAGGAAAC AGCTATGACC CgTRPl on pBluescript

CgKO- TCACTTTCTTCCTCTGTAGTGACCTCGGTAATTTCTAAAACCTCGTCTCC 27

SGSl(-) GTTGTAAAAC GACGGCCAGT

URA75kb(+ ATAGCTAGGT AATTTTAATC TGGGGAGAGA AATGGTGAAC 28

) TTTTTTCAAT CACTATAGGGCGAATTGGGTAC KIURA3 on pBluescript

CTGAAATTGAAGCAGCACCACAAGATATCAATCAACAACCGAATCAAT 29

URA75kb(-) AA AAAGGGAACAAAAGCTGGAGC

r52- GAGAAGTTGGAAGACCAAAGATCAATCCCCTGCATGCACGCAAGCCTA 30

FPfuse(+) CT TCTAAAGGTGAAGAATTATTCACTGG pNB0263 r52-FPfuse(- AGTAATAAATAATGATGCAAATTTTTTATTTGTTTCGGCCAGGAAGCGT 31

) T TTAGTATCGAATCGACAGCAG

ATGTCGAGGG AGTCGAACGA CACAATACAG AGCGATACGG 32

REV3KO(+) TTAGATCATC CACAGGAAAC AGCTATGACC CgTRPl on pBluescript

TTTGAACAGATTGATTATCTCTCAAGTATCTTTCTGCTTTGACACGAGA 33

REV3KO(-) G GTTGTAAAAC GACGGCCAGT

RAD51KO( AAGAGCAGAC GTAGTTATTT GTTAAAGGCC TACTAATTTG 34

+) TTATCGTCAT CACAGGAAAC AGCTATGACC CgTRPl on pBluescript

RAD51KO( AGAATTGAAAGTAAACCTGTGTAAATAAATAGAGACAAGAGACCAAA 35

-) TAC GTTGTAAAAC GACGGCCAGT

RAD52KO( GGAGGTTGCC AAGAACTGCT GAAGGTTCTG GTGGCTTTGG 36

+) TGTGTTGTTG CACAGGAAAC AGCTATGACC CgTRPl on pBluescript

RAD52KO( AGTAATAAATAATGATGCAAATTTTTTATTTGTTTCGGCCAGGAAGCGT 37

-) T GTTGTAAAAC GACGGCCAGT

cgko- AAGGGTTACG TAGAGGAGAA GAGCATATTT CAGGATAAAC 38

rad59(+) AGACAAAATA CACAGGAAAC AGCTATGACC CgTRPl on pBluescript cgko- CTTTAGCATCCTCCAATTTGATAAAAGTCGGCTTGCTATTAGTCGCTGA 39

rad59(-) C GTTGTAAAAC GACGGCCAGT

POL32KO( ATGGATCAAA AGGCGTCATA TTTTATCAAT GAGAAGCTCT 40

+) TCACTGAGGT CACAGGAAAC AGCTATGACC CgTRPl on pBluescript

POL32KO(- TGACGGCGTTTCTTGCTTTTTTGAAGACGACAATGCCCTTTTTACCACG 41

) G GTTGTAAAAC GACGGCCAGT

cgko- GCAGACAATT GACGCAAGTT GTACCTGCTC AGATCCGATA 42

mrell(+) AAACTCGACT CACAGGAAAC AGCTATGAC CgTRPl on pBluescript cgko- AGGAGACTTCCAAGAATATCCGTCTTTGGCGTCCTTGATGCTCTTCCTT 43

mrell(-) T GTTGTAAAAC GACGGCCAGT

CgKO- TAACGAGAGT GCAGGACATA TGCACAAATA ATATATCTCA 44

yku80(+) CACCATAATA CACAGGAAAC AGCTATGACC CgTRPl on pBluescript

CgKO- TTCCCCTACTGTGTTGTTCACCGCGCTTCAATAGCGTTTCTAGGTCCGG 45

yku80(-) G GTTGTAAAAC GACGGCCAGT

CgKO- GATCTTACGG TCTCACTAAC CTCTCTTCAA CTGCTCAATA 46 CgTRPl on pBluescript 5 smll(+) ATTTCCCGCT CACAGGAAAC AGCTATGACC

4 CgKO- CAGAACTAGTGGGAAATGGAAAGAGAAAAGAAAAGAGTATGAAAGG 47

6 smll(-) AACT GTTGTAAAAC GACGGCCAGT

4 PrKO- CACGAAACGT CAACACAATC ATCAAACTCT TTTGCATATT 48

7 ddc2(+) TCTATTATAG GATCTGTTTAGCTTGCCTCGTCCC pNB0132

4 PrKO- TCTTTCCTAAAACGAAAATAATATAAATTATATATAGTTAATATTAAGC 49

8 ddc2(-) A AAACTGGATGGCGGCGGTTAG

check primers

4 50 changes marked with 9 Cgchk(-) GGTCATAGCTGTTTCCTGTG KIURA3 or CgTRPl

5 apnlKOchk( 51

0 +) GCGGC CAAGAAGGAA CCGATTCACG deletion of APN1

5 met25pchk(- 52

1 ) CGAGGCAAGCTAAACAGATC changes marked with KanR

5 URA197chk 53

2 (-) GTACCCAATTCGCCCTATAGTG KIURA3 insertions

5 U A17kbch 54

3 k(+) GACTGGGAAGTTCTGTCGTAG KIURA3 at -17kb

5 URA8kbchk 55

4 (+) CTCAGGAAAATTACTGGCGAAGG KIURA3 at -8kb

5 URA5kbchk 56

5 (+) CGCATCTTCAAACGGCAGCAAG KIURA3 at -5kb

5 URAsfmchk 57

6 (+) cccagcttttgttccctttagtg KIURA3 inside pNB0537

5 URA3kbchk 58

7 v2(+) GTCATTGAGATATGATAGCCTGTTCC KIURA3 at llkb

5 URA197chk 59

8 (+) GCTCCAGCTTTTGTTCCCTTT KIURA3 insertions

5 URAllkbch 60

9 kV2(-) ATGTGCCTGATGAACTAACACAAGG KIURA3 at 15kb

6 URAOkbckv 61

0 2(+) TTCGAAAGCTCTATCATATGGC KIURA3 at CHRI197000

6 ADE2KOch 62

1 k(+) CGCATCTGTTCCTCTATCTTC deletion of ade2-l

6 CANlKOch 63

2 k(+) gcttagcatttgccgttgg deletion of canl-100

6 RAD52KOc 64

3 hk(+) ACTAAATGGTTGAATCGGGTC deletion of RAD52

6 CHRIinsch 65 integration of pNB0537 and 4 V2(+) TTCACTACACCTCGGACATGGATTTG pNB0639

6 CHRIinschk 66

5 (-) CCCTATCAGTGATAGAGAGACGGACG integration of pNB0537

6 URA75kbch 67

6 k(+) GAGGAAAAGATTCATCAACTGGC KIURA3 at -82kb

PrKO- 68

6 EX01chk(+

7 ) CTGAGGTTGACTACTACGAGC Deletion oi EXOl

6 CgKO- 69

8 SGSlchk(+) GAAATGCGAAATGTGAAGGAAGAG Deletion of SGS1

6 REV3KOch 70

9 k(+) GACGAGTGCAGTGCGTCTAG Deletion of REV3

7 POL32KOc 71

0 hk(+) CGGTGTAACTTTCCGACGGAAG Deletion of POL32

7 RAD51KOc 72

1 hk(+) CCACTACCGTTCTTCAACCAATC Deletion of RAD 51

7 rad59kochk( 73

2 +) GTTCTTGTATGTGGCGCTGC Deletion of RAD59

7 mrellkochk 74

3 (+) AGCCAATCATTTCGACCGTC Deletion of MRE11

7 yku80kochk 75

4 (+) CAGTTGGAGGGCGTTAAAAAC Deletion of YKU80

7 smllkochk( ATGTTTAGACCTCGTACATAGG 76 Deletion of SML1 5 +)

7 77

6 ddc2kochk AAGAGTCAGACAGGCTCGC Deletion of DDC2 plasmid construction primers

7 Xhol- 78

7 GAL1(+) GCGGCCTCGAGCAAAAATTCTTACTT GALlpr

7 BamHI- 79

8 GALl(-) GCGGCGGATCCGTTTTTTCTCCTTGACG

7 Spel- 80

9 MAG(+) ccgcgactagtaacaaa ATGAAACTAAAAAGGGAGTATGATG MAGI

8 SacI-MAG(- atattgagctcgttcatgtgcggcgcctaagttctgtcgactta 81

0 ) TTAGGATTTCACGAAATTTTCTTC

Sall- 82

8 ACT1UTR(

1 +) ataatgtcgacgttcatgtgcggccgc TCTGCTTTTGTGCGCGTATG ACT It

Sacl- 83

8 ACT1UTR(-

2 ) cggcggagctc AATTTTTGAAATTTTCGTAGAAAAGGG "

8 MUT- GGTACATACATAAACATACGCGCACAAAAGCAGA ttatta 84

3 (sc)tetR(-) GTCGCCGCTTTCGCACTTTAG sctetR

8 sctetR- ATACTTTAAC GTCAAGGAGA AAAAACTATA AACAAA 85

4 GAL(+) ATGCCGAAAAAAAAACGCAAAGTG tctagattagataaaagtaaag sctetR

8 AT GAAGGCAGAA GAAAATTTCG TGAAATCC GTC GAC GGT GCT GGT TTA ATT AAC 86

5 MAG-YFP(+) TCTAAAG GTG AAG AATTATTCACTG G vYFP

8 GGTACATACATAAACATACGCGCACAAAAGCAGA TTATTA 87

6 ACTlt-YFP(-) TTTGTACAATTCATCCATACCATGG "

8 MAG- AT GAAGGCAGAA GAAAATTTCG TGAAATCC GTC GAC GGT GCT GGT TTA ATT AAC 88

7 (sc)tetR(+) tctagattagataaaagtaaag sctetR

8 ATACTTTAAC GTCAAGGAGA AAAAACTATA AACAAA 89

8 Gal-YFP(+) ATGCCGAAAAAAAAACGCAAAGTG TCTAAAGGTGAAGAATTATTCACTGG vYFP

8 Ascl- 90

9 chrlup(+) tatgcggcggcgcgcc ATTTTGACATATACTGATATGGACCTC CHRI:197000 5' homology

9 Xbal- 91

0 chrlup(-) gcggctctaga TTCAGATATGAGGCCATAAATGGAG "

9 Notl- 92

1 chrlins25 GTGGT GCGGCCGC TTTCAAGTAG TTCACAAAGA CHRI:197000 3' homology

9 Ascl- 93

2 chrlins23(-) ATAAT GGCGCGCC CAATCGCTGG GAATGAGCAA ¹¹

9 XholA- 94

3 ade2(+) gaggactcgagcctagg AAGC I 1 1 1 GACCAGGTTATTATAAAAG ade2-l cassette

9 95

4 Xhol-ade2(-) gaggactcgag CAGGTAATTATTCCTTGCTTCTTG ¹¹

9 96

5 Apal-KIU(+) tatta gggccc ggagacaatc KIURA3 on pBluescript

9 97

6 Hind3-KIU(-) gagga aagctt GCTTATCGCAATGGTTGTAATGG

9 GAL10- I GA I I A I IAAAL I I L I 1 1 GLG 1 LLA 1 LLAAAAAAAAAG 1 AAGAAI 1 1 1 1 G gctagc aacaaa 98

7 MAG1(+) ATGAAACTAAAAAGGGAGTATGATG MAGI

9 tgcgcaactgttgggaagggcgatcggtgcgggcctcttcgctattacgc cccggg 99

8 pBS-ACTl(-) AA 1 1 1 1 1 GAAATTTTCGTAGAAAAGGG

9 sctetR- CTTCTCCTCCAG CTCG CTCTTCACCAGCTG GTTAATTAAACCAGCACCGTCGAC 10

9 FOKI(-) GTCGCCGCTTTCGCACTTTAG 0 sctetR

primers to confirm distances of markers from the 240x array on the telomeric side

1 10

0 1

0 46451(+) tggtcaactcaacgattcttagg genomic DNA

1 10

0 2

1 46241(-) CACTATAGCTTGCTGTATGTCTC " 10

3

42459(+) GAGAAATTGGCTACTTAGGAAGAG

10

4

42393(-) GCTGAATACGATATGGACTAGAG

sequencing primers

10

5

KIU-seql(+) cgttcatggtgacacttttagc "

10

6

KIU-seq2(+) CATCAAATGGTGGTTATTCGTGG "

10

7

KlU-seql(-) GTAAGATGAAGTTGAAGTAGTGTTGC "

10

8

KIU-seq2(-) C 1 C 1 1 1 1 1 CGATGATGTAGTTTCTGG "

* Cassette means promoter, O F, and terminator

Growth, fluorimetry, fluorescence microscopy, and flow cytometry. Yeast strains containing plasmids were grown at 30°C in yeast nitrogen base with appropriate amino acids containing 2% dextrose (SD), except when induction by 2% galactose (SG) or a balance of 5 galactose and raffinose (2% total sugar) was required. Experiments to measure growth rate and fluorescence protein expression were carried out by diluting cells from either a liquid starter culture or fresh plate in appropriate media at a density of 105 or 106 cells/mL. Growth was measured by optical density at 600 nm (OD) at various time points on a Varioskan Flash plate reader (Thermo Scientific). Fluorescence measurements were taken from exponentially growing 10 cells at similar OD by either fluorimetry (Varioskan Flash, Thermo Scientific), flow cytometry (LSR 2, Becton Dickinson), or fluorescence microscopy (Zeiss Axiovert 200M).

Fluctuation analysis. Fluctuation analysis was carried out based on methods described in [45, 46]. Briefly, 12 parallel cultures were grown without agitation from low density (10,000 cells/mL) to saturation for 3-4 days in SG for induction. Small (20 uL) and large (0.5-1 mL)

15 volume cultures were used to measure high and low mutation rates, respectively. To convert OD to cell density, a calibration factor was determined by growing 48 parallel cultures to saturation in the same conditions and plating dilutions on YPD medium. After determining the OD, the entire culture was plated on 30mm diameter plates to facilitate analysis of many cultures. Selection plates consisted of SD media and 1 g/L 5'FOA (USBiological) for klura3

20 mutants or SD media without arginine and 600 mg/L canavanine (Sigma) for canl mutants. Selection on FOA and canavanine plates required 2 and 4 days of growth, respectively. Plates were imaged at 4x magnification and colony number was scored using custom image analysis software written in MATLAB (Mathworks). A maximum likelihood estimate of mutation rate for the distribution of mutants in each culture was found using a MATLAB implementation of the MSS equation [47] . 95% confidence intervals were calculated using equation 3 in [48].

Observation of Rad52-CFP foci: Cells expressing Rad52-CFP were grown as described above, harvested at an OD between 1 and 2, and imaged on an optical microscope. HO induction was accomplished by overnight growth in 2% raffinose followed by 8 hour induction in 2% galactose. Foci were counted by observing the change in brightness across a z-stack of images for the brightest 9 pixels in a cell. This change was used as a threshold which was calibrated such that the HO induced fractions of cells with dots matched those in [25].

Determination of cell cycle distribution: Cells were grown overnight to an OD between 0.5 and 0.8 in SG media without leucine to induce mutators and select for plasmids. Cells were collected, fixed, and DNA was stained with S YTOX green (Invitrogen) according to [49] . Flow cytometry was performed on a BD Accuri C6 flow cytometer (Becton Dickinson).

Nucleotide and protein sequences of reagents described herein (SEQ ID NO: 109 through SEQ ID NO: 122):

Magl-sctetR Nucleotide (SEQ ID NO: 109):

atgaaactaaaaagggagtatgatgagttaataaaagcagacgctgttaaggaaatagcaaaagaattagggtctcgacctctagaggt tgctcttcctgagaaatatattgctagacatgaagaaaagttcaatatggcttgcgaacacattttagagaaagatccatcactttttcccata cttaagaataatgaatttacgttgtacttgaaggagactcaagtccctaatacactcgaagattattttattaggcttgcaagcacaattttgtc tcaacagatcagtggccaagcagctgaaagcatcaaggcaagggttgtcagtctttatggcggtgcatttcctgattacaaaatccttttc gaagacttcaaagacccagcaaaatgtgcagaaatcgcaaaatgtggattgagtaaaaggaaaatgatatatctagagtctcttgctgtc tattttactgaaaaatataaggatatcgaaaagctcttcggtcaaaaagataatgatgaggaagtgattgaaagtttagttacgaatgtaaa aggtataggcccatggagtgccaaaatgttcttgatctccggattgaaaagaatggatgtatttgctcctgaagatctaggtattgctaggg gtttttcaaaatacctttcagataagccagaattggaaaaagaattaatgcgtgaaagaaaagtagttaaaaagagtaagattaagcataa gaaatacaactggaaaatatatgacgacgacataatggaaaaatgctctgaaacattttctccgtataggtctgtgtttatgttcatactttgg aggctcgcgagcacaaatacagatgccatgatgaaggcagaagaaaatttcgtgaaatccgtcgacggtgctggtttaattaactctag attagataaaagtaaagtgattaacagcgcattagagctgcttaatgaggtcggaatcgaaggtttaacaacccgtaaactcgcccagaa gcttggtgtagagcagcctacattgtattggcatgtaaaaaataagcgggctttgctcgacgccttagccattgagatgttagataggcac catactcacttttgccctttagaaggggaaagctggcaagattttttacgtaataacgctaaaagttttagatgtgctttactaagtcatcgcg atggagcaaaagtacatttaggtacacggcctacagaaaaacagtatgaaactctcgaaaatcaattagcctttttatgccaacaaggtttt tcactagagaacgcgttatatgcactcagcgctgtggggcattttaccttaggttgcgtattggaagatcaagagcatcaagtcgctaaag aagaaagggaaacacctactactgatagtatgccgccattattacgacaagctatcgaattatttgatcaccaaggtgcagagccagcct tcttattcggccttgaattgatcatatgcggattagaaaaacaacttaaatgtgaaagtggggactcaggaggcggtggatccggtggag gcggttctggcggtggaggctctggaggtggcggatcgggcggaggtggctctagatctcgtctggacaagagcaaagtcataaact ctgctctggaattactcaatgaagtcggtatcgaaggcctgacgacaaggaaactcgctcaaaagctgggagttgagcagcctaccct gtactggcacgtgaagaacaagcgggccctgctcgatgccctggcaatcgagatgctggacaggcatcatacccacttctgccccctg gaaggcgagtcatggcaagactttctgcggaacaacgccaagtcattccgctgtgctctcctctcacatcgcgacggggctaaagtgc atctcggcacccgcccaacagagaaacagtacgaaaccctggaaaatcagctcgcgttcctgtgtcagcaaggcttctccctggagaa cgcactgtacgctctgtccgccgtgggccactttacactgggctgcgtattggaggatcaggagcatcaagtagcaaaagaggaaaga gagacacctaccaccgattctatgcccccacttctgagacaagcaattgagctgttcgaccatcagggagccgaacctgccttccttttc ggcctggaactaatcatatgtggcctggagaaacagctaaagtgcgaaagcggcgactaa

Magl-sctetR Protein (SEQ ID NO: 110)

MKLKREYDELIKADAVKEIAKELGSRPLEVALPEKYIARHEEKFNMACEHILEKDPSL FPILKNNEFTLYLKETQVPNTLEDYFIRLASTILSQQISGQAAESIKARVVSLYGGAFPD YKILFEDFKDPAKCAEIAKCGLSKRKMIYLESLAVYFTEKYKDIEKLFGQKDNDEEVI ESLVTNVKGIGPWSAKMFLISGLKRMDVFAPEDLGIARGFSKYLSDKPELEKELMRE RKVVKKSKIKHKKYNWKIYDDDIMEKCSETFSPYRSVFMFILWRLASTNTDAMMKA EENFVKSVDGAGLINSRLDKSKVINSALELLNEVGIEGLTTRKLAQKLGVEQPTLYW HVKNKRALLDALAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRDGAKV HLGTRPTEKQYETLENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEHQVAKE ERETPTTDSMPPLLRQAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGDSGGGGSGG GGSGGGGSGGGGSGGGGSRSRLDKSKVINSALELLNEVGIEGLTTRKLAQKLGVEQP TLYWHVKNKRALLDALAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRD GAKVHLGTRPTEKQYETLENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEH QVAKEERETPTTDSMPPLLRQAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGD

CDG-sctetR Nucleotide (SEQ ID NO: 111)

atgccgaaaaaaaaacgcaaagtgtttggagagagctggaagaagcacctcagcggggagttcgggaaaccgtattttatcaagctaa tgggatttgttgcagaagaaagaaagcattacactgtttatccacccccacaccaagtcttcacctggacccagatgtgtgacataaaag atgtgaaggttgtcatcctgggacaggatccatatcatggacctaatcaagctcacgggctctgctttagtgttcaaaggcctgttccgcct ccgcccagtttggagaacatttataaagagttgtctacagacatagaggattttgttcatcctggccatggagatttatctgggtgggccaa gcaaggtgttctccttctcaacgctgtcctcacggttcgtgcccatcaagccaactctcataaggagcgaggctgggagcagttcactga tgcagttgtgtcctggctaaatcagaactcgaatggccttgttttcttgctctggggctcttatgctcagaagaagggcagtgccattgata ggaagcggcaccatgtactacagacggctcatccctcccctttgtcagtgtatagagggttctttggatgtagacacttttcaaagaccga tgagctgctgcagaagtctggcaagaagcccattgactggaaggagctggtcgacggtgctggtttaattaactctagattagataaaag taaagtgattaacagcgcattagagctgcttaatgaggtcggaatcgaaggtttaacaacccgtaaactcgcccagaagcttggtgtaga gcagcctacattgtattggcatgtaaaaaataagcgggctttgctcgacgccttagccattgagatgttagataggcaccatactcactttt gccctttagaaggggaaagctggcaagattttttacgtaataacgctaaaagttttagatgtgctttactaagtcatcgcgatggagcaaaa gtacatttaggtacacggcctacagaaaaacagtatgaaactctcgaaaatcaattagcctttttatgccaacaaggtttttcactagagaa cgcgttatatgcactcagcgctgtggggcattttaccttaggttgcgtattggaagatcaagagcatcaagtcgctaaagaagaaaggga aacacctactactgatagtatgccgccattattacgacaagctatcgaattatttgatcaccaaggtgcagagccagccttcttattcggcc ttgaattgatcatatgcggattagaaaaacaacttaaatgtgaaagtggggactcaggaggcggtggatccggtggaggcggttctggc ggtggaggctctggaggtggcggatcgggcggaggtggctctagatctcgtctggacaagagcaaagtcataaactctgctctggaat tactcaatgaagtcggtatcgaaggcctgacgacaaggaaactcgctcaaaagctgggagttgagcagcctaccctgtactggcacgt gaagaacaagcgggccctgctcgatgccctggcaatcgagatgctggacaggcatcatacccacttctgccccctggaaggcgagtc atggcaagactttctgcggaacaacgccaagtcattccgctgtgctctcctctcacatcgcgacggggctaaagtgcatctcggcaccc gcccaacagagaaacagtacgaaaccctggaaaatcagctcgcgttcctgtgtcagcaaggcttctccctggagaacgcactgtacgc tctgtccgccgtgggccactttacactgggctgcgtattggaggatcaggagcatcaagtagcaaaagaggaaagagagacacctac caccgattctatgcccccacttctgagacaagcaattgagctgttcgaccatcagggagccgaacctgccttccttttcggcctggaacta atcatatgtggcctggagaaacagctaaagtgcgaaagcggcgactaa

Protein CDG-sctetR (SEQ ID NO: 112)

MPKK RKVFGESWKKHLSGEFGKPYFIKLMGFVAEERKHYTVYPPPHQVFTWTQM CDIKDVKVVILGQDPYHGPNQAHGLCFSVQRPVPPPPSLENIYKELSTDIEDFVHPGH GDLSGWAKQGVLLLNAVLTVRAHQANSHKERGWEQFTDAVVSWLNQNSNGLVFL LWGSYAQKKGSAIDRKRHHVLQTAHPSPLSVYRGFFGCRHFSKTDELLQKSGKKPID WKELVDGAGLINSRLDKSKVINSALELLNEVGIEGLTTRKLAQKLGVEQPTLYWHVK NKRALLDALAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRDGAKVHLG TRPTEKQYETLENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEHQVAKEERE TPTTDSMPPLLRQAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGDSGGGGSGGGGS GGGGSGGGGSGGGGSRSRLDKSKVINSALELLNEVGIEGLTTRKLAQKLGVEQPTLY WHVKNKRALLDALAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRDGAK VHLGTRPTEKQYETLENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEHQVA KEERETPTTDSMPPLLRQAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGD

sctetR-Fokl Nucleotide (SEQ ID NO: 113)

atgccgaaaaaaaaacgcaaagtgtctagattagataaaagtaaagtgattaacagcgcattagagctgcttaatgaggtcggaatcga aggtttaacaacccgtaaactcgcccagaagcttggtgtagagcagcctacattgtattggcatgtaaaaaataagcgggctttgctcga cgccttagccattgagatgttagataggcaccatactcacttttgccctttagaaggggaaagctggcaagattttttacgtaataacgcta aaagttttagatgtgctttactaagtcatcgcgatggagcaaaagtacatttaggtacacggcctacagaaaaacagtatgaaactctcga aaatcaattagcctttttatgccaacaaggtttttcactagagaacgcgttatatgcactcagcgctgtggggcattttaccttaggttgcgta ttggaagatcaagagcatcaagtcgctaaagaagaaagggaaacacctactactgatagtatgccgccattattacgacaagctatcga attatttgatcaccaaggtgcagagccagccttcttattcggccttgaattgatcatatgcggattagaaaaacaacttaaatgtgaaagtg gggactcaggaggcggtggatccggtggaggcggttctggcggtggaggctctggaggtggcggatcgggcggaggtggctctag atctcgtctggacaagagcaaagtcataaactctgctctggaattactcaatgaagtcggtatcgaaggcctgacgacaaggaaactcg ctcaaaagctgggagttgagcagcctaccctgtactggcacgtgaagaacaagcgggccctgctcgatgccctggcaatcgagatgc tggacaggcatcatacccacttctgccccctggaaggcgagtcatggcaagactttctgcggaacaacgccaagtcattccgctgtgct ctcctctcacatcgcgacggggctaaagtgcatctcggcacccgcccaacagagaaacagtacgaaaccctggaaaatcagctcgcg ttcctgtgtcagcaaggcttctccctggagaacgcactgtacgctctgtccgccgtgggccactttacactgggctgcgtattggaggat caggagcatcaagtagcaaaagaggaaagagagacacctaccaccgattctatgcccccacttctgagacaagcaattgagctgttcg accatcagggagccgaacctgccttccttttcggcctggaactaatcatatgtggcctggagaaacagctaaagtgcgaaagcggcga cgtcgacggtgctggtttaattaaccagctggtgaagagcgagctggaggagaagaagtccgagctgcggcacaagctgaagtacgt gccccacgagtacatcgagctgatcgagatcgccaggaacagcacccaggaccgcatcctggagatgaaggtgatggagttcttcat gaaggtgtacggctacaggggaaagcacctgggcggaagcagaaagcctgacggcgccatctatacagtgggcagccccatcgatt acggcgtgatcgtggacacaaaggcctacagcggcggctacaatctgcctatcggccaggccgacgagatggagagatacgtgga ggagaaccagacccggaataagcacctcaaccccaacgagtggtggaaggtgtaccctagcagcgtgaccgagttcaagttcctgtt cgtgagcggccacttcaagggcaactacaaggcccagctgaccaggctgaaccacatcaccaactgcaatggcgccgtgctgagcg tggaggagctgctgatcggcggcgagatgatcaaagccggcaccctgacactggaggaggtgcggcgcaagttcaacaacggcga gatcaacttcagatcttgataa

sctetR-Fokl Protein (SEQ ID NO: 114)

MPKKKRKVSRLDKSKVINSALELLNEVGIEGLTTRKLAQKLGVEQPTLYWHVKNKR ALLDALAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRDGAKVHLGTRPT EKQYETLENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEHQVAKEERETPTT DSMPPLLRQAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGDSGGGGSGGGGSGGG GSGGGGSGGGGSRSRLDKSKVINSALELLNEVGIEGLTTRKLAQKLGVEQPTLYWHV KNKRALLDALAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRDGAKVHL GTRPTEKQYETLENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEHQVAKEER ETPTTDSMPPLLRQAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGDVDGAGLINQL VKSELEEKKSELRHKLKYVPHEYIELIEIARNSTQDRILEMKVMEFFMKVYGYRGKH LGGSRKPDGAIYTVGSPIDYGVIVDTKAYSGGYNLPIGQADEMERYVEENQTRNKHL NPNEWWKVYPSSVTEFKFLFVSGHFKGNYKAQLTRLNHITNCNGAVLSVEELLIGGE MIKAGTLTLEEVRRKFNNGEINFRS

sctetR-scFokl Nucleotide (SEQ ID NO: 115)

atgccgaaaaaaaaacgcaaagtgtctagattagataaaagtaaagtgattaacagcgcattagagctgcttaatgaggtcggaatcga aggtttaacaacccgtaaactcgcccagaagcttggtgtagagcagcctacattgtattggcatgtaaaaaataagcgggctttgctcga cgccttagccattgagatgttagataggcaccatactcacttttgccctttagaaggggaaagctggcaagattttttacgtaataacgcta aaagttttagatgtgctttactaagtcatcgcgatggagcaaaagtacatttaggtacacggcctacagaaaaacagtatgaaactctcga aaatcaattagcctttttatgccaacaaggtttttcactagagaacgcgttatatgcactcagcgctgtggggcattttaccttaggttgcgta ttggaagatcaagagcatcaagtcgctaaagaagaaagggaaacacctactactgatagtatgccgccattattacgacaagctatcga attatttgatcaccaaggtgcagagccagccttcttattcggccttgaattgatcatatgcggattagaaaaacaacttaaatgtgaaagtg gggactcaggaggcggtggatccggtggaggcggttctggcggtggaggctctggaggtggcggatcgggcggaggtggctctag atctcgtctggacaagagcaaagtcataaactctgctctggaattactcaatgaagtcggtatcgaaggcctgacgacaaggaaactcg ctcaaaagctgggagttgagcagcctaccctgtactggcacgtgaagaacaagcgggccctgctcgatgccctggcaatcgagatgc tggacaggcatcatacccacttctgccccctggaaggcgagtcatggcaagactttctgcggaacaacgccaagtcattccgctgtgct ctcctctcacatcgcgacggggctaaagtgcatctcggcacccgcccaacagagaaacagtacgaaaccctggaaaatcagctcgcg ttcctgtgtcagcaaggcttctccctggagaacgcactgtacgctctgtccgccgtgggccactttacactgggctgcgtattggaggat caggagcatcaagtagcaaaagaggaaagagagacacctaccaccgattctatgcccccacttctgagacaagcaattgagctgttcg accatcagggagccgaacctgccttccttttcggcctggaactaatcatatgtggcctggagaaacagctaaagtgcgaaagcggcga cgtcgacggtgctggtttaattaacagctggttaaaagcgaacttgaagagaaaaagtccgagctgcgccataaacttaagtatgtgccg cacgaatacattgagctgattgaaatcgctcgcaatagcactcaagatagaatccttgaaatgaaggttatggagtttttcatgaaggtgta tggttacagaggcaaacatctgggtggctctcgtaagcctgatggtgcaatttatacagttggctcaccgattgattacggtgttatcgtgg ataccaaagcttattctggaggttacaacctgccaattggtcaggcagatgaaatgcaacgctatgttgaagagaatcagaccagaaac aaacacatcaatccaaacgaatggtggaaggtttacccttcttcagtgactgagtttaaattcctgtttgtgagtggccatttcaagggaaa ctacaaggctcaactgactagacttaaccacattacaaattgtaacggtgcagttctttctgtggaagagctgcttattggcggagaaatg atcaaagctggtaccctgactcttgaagaggtgcgtcgcaagtttaataacggtgaaatcaattttccatggggtggcggaggtagtggc ggaggtggctctggaggtggcggatctggtggcggaggttcaggcggaggtggcagcggaggtggcggaagtggtggcggaggt tccggcggaggtggctctggaggtggcggatcaggtggcggaggtagcggcggaggtggcagtggaggtggcggttccggtggc ggaggttctggcggaggtggctccggaggtggtttaattaaccagctggtgaagagcgagctggaggagaagaagtccgagctgcg gcacaagctgaagtacgtgccccacgagtacatcgagctgatcgagatcgccaggaacagcacccaggaccgcatcctggagatga aggtgatggagttcttcatgaaggtgtacggctacaggggaaagcacctgggcggaagcagaaagcctgacggcgccatctatacag tgggcagccccatcgattacggcgtgatcgtggacacaaaggcctacagcggcggctacaatctgcctatcggccaggccgacgag atggagagatacgtggaggagaaccagacccggaataagcacctcaaccccaacgagtggtggaaggtgtaccctagcagcgtga ccgagttcaagttcctgttcgtgagcggccacttcaagggcaactacaaggcccagctgaccaggctgaaccacatcaccaactgcaa tggcgccgtgctgagcgtggaggagctgctgatcggcggcgagatgatcaaagccggcaccctgacactggaggaggtgcggcgc aagttcaacaacggcgagatcaacttcagatcttgataa

sctetR-scFokl Protein (SEQ ID NO: 116) MPKKKRKVSRLDKSKVINSALELLNEVGIEGLTTRKLAQKLGVEQPTLYWHV NKR ALLDALAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRDGAKVHLGTRPT EKQYETLENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEHQVAKEERETPTT DSMPPLLRQAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGDSGGGGSGGGGSGGG GSGGGGSGGGGSRSRLDKSKVINSALELLNEVGIEGLTTRKLAQKLGVEQPTLYWHV KNKRALLDALAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRDGAKVHL GTRPTEKQYETLENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEHQVAKEER ETPTTDSMPPLLRQAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGDVDGAGLINQL VKSELEEKKSELRHKLKYVPHEYIELIEIARNSTQDRILEMKVMEFFMKVYGYRGKH LGGSRKPDGAIYTVGSPIDYGVIVDTKAYSGGYNLPIGQADEMQRYVEENQTRNKHI NPNEWW VYPSSVTEFKFLFVSGHFKGNYKAQLTRLNHITNCNGAVLSVEELLIGGE MIKAGTLTLEEVRRKFNNGEINFPWGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSG GGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGLINQLVKSELEEK KSELRHKLKYVPHEYIELIEIARNSTQDRILEMKVMEFFMKVYGYRGKHLGGSRKPD GAIYTVGSPIDYGVIVDTKAYSGGYNLPIGQADEMERYVEENQTRNKHLNPNEWWK VYPSSVTEFKFLFVSGHFKGNYKAQLTRLNHITNCNGAVLSVEELLIGGEMIKAGTLT LEEVRRKFNNGEINFRS

sctetR-Fokl (Sharkey variant: http://www.ncbi.nlm.nih.gov/pubmed/20447404) Nucleotide (SEQ ID NO: 117)

atgccgaaaaaaaaacgcaaagtgtctagattagataaaagtaaagtgattaacagcgcattagagctgcttaatgaggtcggaatcga aggtttaacaacccgtaaactcgcccagaagcttggtgtagagcagcctacattgtattggcatgtaaaaaataagcgggctttgctcga cgccttagccattgagatgttagataggcaccatactcacttttgccctttagaaggggaaagctggcaagattttttacgtaataacgcta aaagttttagatgtgctttactaagtcatcgcgatggagcaaaagtacatttaggtacacggcctacagaaaaacagtatgaaactctcga aaatcaattagcctttttatgccaacaaggtttttcactagagaacgcgttatatgcactcagcgctgtggggcattttaccttaggttgcgta ttggaagatcaagagcatcaagtcgctaaagaagaaagggaaacacctactactgatagtatgccgccattattacgacaagctatcga attatttgatcaccaaggtgcagagccagccttcttattcggccttgaattgatcatatgcggattagaaaaacaacttaaatgtgaaagtg gggactcaggaggcggtggatccggtggaggcggttctggcggtggaggctctggaggtggcggatcgggcggaggtggctctag atctcgtctggacaagagcaaagtcataaactctgctctggaattactcaatgaagtcggtatcgaaggcctgacgacaaggaaactcg ctcaaaagctgggagttgagcagcctaccctgtactggcacgtgaagaacaagcgggccctgctcgatgccctggcaatcgagatgc tggacaggcatcatacccacttctgccccctggaaggcgagtcatggcaagactttctgcggaacaacgccaagtcattccgctgtgct ctcctctcacatcgcgacggggctaaagtgcatctcggcacccgcccaacagagaaacagtacgaaaccctggaaaatcagctcgcg ttcctgtgtcagcaaggcttctccctggagaacgcactgtacgctctgtccgccgtgggccactttacactgggctgcgtattggaggat caggagcatcaagtagcaaaagaggaaagagagacacctaccaccgattctatgcccccacttctgagacaagcaattgagctgttcg accatcagggagccgaacctgccttccttttcggcctggaactaatcatatgtggcctggagaaacagctaaagtgcgaaagcggcga cgtcgacggtgctggtttaattaaccagctggtgaagagcgagctggaggagaagaagtccgagctgcggcacaagctgaagtacgt gccccacgagtacatcgagctgatcgagatcgccaggaacccaacccaggaccgcatcctggagatgaaggtgatggagttcttgat gaaggtgtacggctacaggggagaacacctgggcggaagcagaaagcctgacggcgccatctatacagtgggcagccccatcgatt acggcgtgatcgtggacacaaaggcctacagcggcggctacaatctgcctatcggccatgccgacgagatggagagatacgtggag gagaaccagacccggaataagcacctcaaccccaacgagtggtggaaggtgtaccctagcagcgtgaccgagttcaagttcctgttc gtgagcggctatttcaagggcgattacaaggcccagctgaccaggctgaaccacatcaccaactgcaatggcgccgtgctgagcgtg gaggagctgctgatcggcggcgagatgatccaagccggcaccctgacactggaggaggtgcggcgcaagttcaacaacggcgag atcaacttcagatcttgataa

sctetR-Fokl (Sharkey variant: http://www.ncbi.nlm.nih.gov/pubmed/20447404) Protein (SEQ ID NO: 118) MPKKKRKVSRLDKSKVINSALELLNEVGIEGLTTR LAQ LGVEQPTLYWHVKNKRAL LDALAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRDGAKVHLGTRPTE Q YETLENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEHQVA EERETPTTDSMPP LLRQAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGDSGGGGSGGGGSGGGGSGGGG SGGGGSRSRLDKSKVINSALELLNEVGIEGLTTRKLAQKLGVEQPTLYWHVKNKRALL DALAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRDGAKVHLGTRPTEKQY ETLENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEHQVAKEERETPTTDSMPPL LRQAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGDVDGAGLINQLVKSELEEKKSEL RHKLKYVPHEYIELIEIARNPTQDRILEMKVMEFLMKVYGYRGEHLGGSRKPDGAIYTV GSPIDYGVIVDTKAYSGGYNLPIGHADEMERYVEENQTRNKHLNPNEWWKVYPSSVTE FKFLFVSGYFKGDYKAQLTRLNHITNCNGAVLSVEELLIGGEMIQAGTLTLEEVRRKFN NGEINFRS

AlkA-sctetR Nucleotide: (SEQ ID NO: 119)

atgtataccctgaactggcagccgccgtatgactggtcgtggatgttgggatttctcgccgcccgtgcggtgagcagtgtggaaacggtcg cggacagttattatgcccgtagtctggcggtgggcgaatatcgcggcgtggtgactgctattccggatatagcccgccatactctgcacata aatttaagtgcaggtttagaacctgttgccgcagagtgtctggcgaaaatgagccgcctgtttgatctgcaatgtaacccacaaattgttaacg gtgcgttgggcaggttaggcgcggcgcggcccggattgcgtttacccggctgtgttgatgcttttgagcagggcgtgcgggcgattttagg ccaactggtgagcgtggcgatggcggcaaaattgaccgccagagtggcacagctttatggcgaacggctggatgattttccggagtatatc tgcttcccgacgcctcagcggctggcagcagcagacccgcaggcattaaaagcgttaggtatgccgttgaaacgggcagaggcgctgat tcatctggcaaatgcggcgctggagggcaccttaccaatgacaataccgggcgatgtggagcaggcgatgaaaacgctgcaaacttttcc gggtatcgggcgctggacggcgaattattttgctttgcgtggctggcaggcgaaagatgttttcctgccggatgattatctgattaaacagcg atttccgggaatgacaccggcgcaaatccgccgttatgccgagcgctggaagccctggcgttcttatgcgctgttgcatatctggtatacgg aaggctggcaaccagacgaagcagtcgacggtgctggtttaattaactctagattagataaaagtaaagtgattaacagcgcattagagctg cttaatgaggtcggaatcgaaggtttaacaacccgtaaactcgcccagaagcttggtgtagagcagcctacattgtattggcatgtaaaaaat aagcgggctttgctcgacgccttagccattgagatgttagataggcaccatactcacttttgccctttagaaggggaaagctggcaagattttt tacgtaataacgctaaaagttttagatgtgctttactaagtcatcgcgatggagcaaaagtacatttaggtacacggcctacagaaaaacagta tgaaactctcgaaaatcaattagcctttttatgccaacaaggtttttcactagagaacgcgttatatgcactcagcgctgtggggcattttacctt aggttgcgtattggaagatcaagagcatcaagtcgctaaagaagaaagggaaacacctactactgatagtatgccgccattattacgacaa gctatcgaattatttgatcaccaaggtgcagagccagccttcttattcggccttgaattgatcatatgcggattagaaaaacaacttaaatgtga aagtggggactcaggaggcggtggatccggtggaggcggttctggcggtggaggctctggaggtggcggatcgggcggaggtggctc tagatctcgtctggacaagagcaaagtcataaactctgctctggaattactcaatgaagtcggtatcgaaggcctgacgacaaggaaactcg ctcaaaagctgggagttgagcagcctaccctgtactggcacgtgaagaacaagcgggccctgctcgatgccctggcaatcgagatgctg gacaggcatcatacccacttctgccccctggaaggcgagtcatggcaagactttctgcggaacaacgccaagtcattccgctgtgctctcct ctcacatcgcgacggggctaaagtgcatctcggcacccgcccaacagagaaacagtacgaaaccctggaaaatcagctcgcgttcctgt gtcagcaaggcttctccctggagaacgcactgtacgctctgtccgccgtgggccactttacactgggctgcgtattggaggatcaggagca tcaagtagcaaaagaggaaagagagacacctaccaccgattctatgcccccacttctgagacaagcaattgagctgttcgaccatcaggg agccgaacctgccttccttttcggcctggaactaatcatatgtggcctggagaaacagctaaagtgcgaaagcggcgactaa

AlkA-sctetR Protein (SEQ ID NO: 120)

MYTLNWQPPYDWSWMLGFLAARAVSSVETVADSYYARSLAVGEYRGVVTAIP DIARHTLHINLSAGLEPVAAECLAKMSRLFDLQCNPQIVNGALGRLGAARPGLRLPGCV DAFEQGVRAILGQLVSVAMAAKLTARVAQLYGERLDDFPEYICFPTPQRLAAADPQAL KALGMPLKRAEALIHLANAALEGTLPMTIPGDVEQAMKTLQTFPGIGRWTANYFALRG WQAKDVFLPDDYLIKQRFPGMTPAQIRRYAERWKPWRSYALLHIWYTEGWQPDEAVD GAGLINSRLDKSKVINSALELLNEVGIEGLTTRKLAQKLGVEQPTLYWHVKNKRALLD ALAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRDGAKVHLGTRPTEKQYE TLENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEHQVAKEERETPTTDSMPPLL RQAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGDSGGGGSGGGGSGGGGSGGGGSG GGGSRSRLD SKVINSALELLNEVGIEGLTTRKLAQKLGVEQPTLYWHVKNKRALLDA LAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRDGAKVHLGTRPTEKQYET LENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEHQVAKEERETPTTDSMPPLLR QAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGD

MPG-sctetR (AAG1 from H. sapiens is more commonly known as MPG) Nucleotide (SEQ ID NO: 121)

atggtcacccccgctttgcagatgaagaaaccaaagcagttttgccgacggatggggcaaaagaagcagcgaccagctagagcagg gcagccacacagctcgtccgacgcagcccaggcacctgcagagcagccacacagctcgtccgatgcagccgaggcaccttgcccc agggagcgctgcttgggaccgcccaccactccgggcccataccgcagcatctatttctcaagcccaaagggccaccttacccgactg gggttggagttcttcgaccagccggcagtccccctggcccgggcatttctgggacaggtcctagtccggcgacttcctaatggcacag aactccgaggccgcatcgtggagaccgaggcatacctgtggccagaggatgaaccggcccactcaaggggtggccgggagaccc cccgcaaccgaggcatgttcatgaagccggggaccctgtacgtgtacatcatttacggcatgtacttctgcatgaacatctccagccag ggggacggggcttgcgtcttgctgcgagcactggagcccctggaaggtctggagaccatgcgtcacgttcgcagcaccctccggaa aggcaccgccagccgtgtcctcaaggaccgcgagctctgcagtggcccctccaagctgtgccaggccctggccatcaacaagagctt tgacgagagggacctggcacaggatgaagctgtatggctggagcgtggtcccctggagcccagtgagccggctgtagtggcagcag cccgggtgggcgtcggccatgcaggggagtgggcccggaaacccctccgcttctatgtccggggcagcccctgggtcagtgtggtc gacagagtggctgagcaggacacacaggccgtcgacggtgctggtttaattaactctagattagataaaagtaaagtgattaacagcgc attagagctgcttaatgaggtcggaatcgaaggtttaacaacccgtaaactcgcccagaagcttggtgtagagcagcctacattgtattg gcatgtaaaaaataagcgggctttgctcgacgccttagccattgagatgttagataggcaccatactcacttttgccctttagaaggggaa agctggcaagattttttacgtaataacgctaaaagttttagatgtgctttactaagtcatcgcgatggagcaaaagtacatttaggtacacgg cctacagaaaaacagtatgaaactctcgaaaatcaattagcctttttatgccaacaaggtttttcactagagaacgcgttatatgcactcag cgctgtggggcattttaccttaggttgcgtattggaagatcaagagcatcaagtcgctaaagaagaaagggaaacacctactactgatag tatgccgccattattacgacaagctatcgaattatttgatcaccaaggtgcagagccagccttcttattcggccttgaattgatcatatgcgg attagaaaaacaacttaaatgtgaaagtggggactcaggaggcggtggatccggtggaggcggttctggcggtggaggctctggagg tggcggatcgggcggaggtggctctagatctcgtctggacaagagcaaagtcataaactctgctctggaattactcaatgaagtcggtat cgaaggcctgacgacaaggaaactcgctcaaaagctgggagttgagcagcctaccctgtactggcacgtgaagaacaagcgggccc tgctcgatgccctggcaatcgagatgctggacaggcatcatacccacttctgccccctggaaggcgagtcatggcaagactttctgcgg aacaacgccaagtcattccgctgtgctctcctctcacatcgcgacggggctaaagtgcatctcggcacccgcccaacagagaaacagt acgaaaccctggaaaatcagctcgcgttcctgtgtcagcaaggcttctccctggagaacgcactgtacgctctgtccgccgtgggccac tttacactgggctgcgtattggaggatcaggagcatcaagtagcaaaagaggaaagagagacacctaccaccgattctatgcccccac ttctgagacaagcaattgagctgttcgaccatcagggagccgaacctgccttccttttcggcctggaactaatcatatgtggcctggaga aacagctaaagtgcgaaagcggcgactaa

MPG-sctetR (AAG1 from H. sapiens is more commonly known as MPG) Protein (SEQ ID NO: 122)

MVTPALQMKKPKQFCRRMGQKKQRPARAGQPHSSSDAAQAPAEQPHSSSDAAEAPCP RERCLGPPTTPGPYRSIYFSSPKGHLTRLGLEFFDQPAVPLARAFLGQVLVRRLPNGTEL RGRIVETEAYLWPEDEPAHSRGGRETPRNRGMFMKPGTLYVYIIYGMYFCMNISSQGD GACVLLRALEPLEGLETMRHVRSTLRKGTASRVLKDRELCSGPSKLCQALAINKSFDER DLAQDEAVWLERGPLEPSEPAVVAAARVGVGHAGEWARKPLRFYVRGSPWVSVVDR VAEQDTQAVDGAGLINSRLDKSKVINSALELLNEVGIEGLTTRKLAQKLGVEQPTLYW HVKNKRALLDALAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRDGAKVH LGTRPTEKQYETLENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEHQVAKEER ETPTTDSMPPLLRQAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGDSGGGGSGGGGS GGGGSGGGGSGGGGSRSRLDKSKVINSALELLNEVGIEGLTTRKLAQKLGVEQPTLYW HVKNKRALLDALAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRDGAKVH LGTRPTEKQYETLENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEHQVAKEER ETPTTDSMPPLLRQAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGD

Example 1: DNA binding site arrays targeted Magi and Fokl to mimic natural DNA damage.

To probe the mutagenic fate of localized DNA damage we have developed a strategy to generate and target damage in a controllable and repeatable way. An example of a strategy taken was to fuse an enzymatic moiety capable of introducing damage to a DNA binding domain and targeting the resulting chimeric protein to a particular genomic location by introducing an array of cognate DNA binding sites. The enzymes were chosen such that the targeted damage would reflect two common types of DNA damage: base pair modification and single- or double-strand breaks.

A large class of DNA damage involves deamination, oxidation, alkylation, and other chemical modifications to bases. This damage is largely repaired by Base Excision Repair (BER), in which the first step is removal of the damaged base by a DNA glycosylase to leave an abasic site. To mimic this kind of damage, the yeast 3-methyladenine glycosylase Maglp was targeted. Maglp is thought to remove normal bases upon over-expression, leading to an increased mutation rate [18] because high levels of abasic sites overwhelm BER and cause fork stalling that requires TLS [19]. Another class of DNA damage involves cutting of the DNA backbone leading to single- and double-strand breaks. To better mimic natural "dirty" SSBs and DSBs, the promiscuous blunt-end endonuclease Fokl was chosen. Native Fokl is prohibited from making SSBs by its binding domain, which sequesters the nuclease domain until dimerization with another Fokl monomer [20] However, as only the nuclease domain of Fokl was used, it likely produces both SSBs and DSBs.

Each enzyme was fused to the tet repressor (tetR), which recognizes a 19 bp tet operator (tetO) sequence. Because normal tetR forms a dimer, the fusion may result in impaired enzymatic or binding activity due to loss of conformational freedom or improper dimerization in the case of Fokl. The use of single chain (sc)tetR - a tandem repeat of tetR connected via a peptide linker [21] - eliminated this problem. N-terminal sctetR-Fokl and C-terminal Magi- sctetR fusions were expressed from the galactose-inducible GAL1 promoter. Magl-sctetR's ability to bind tetO sites was confirmed by showing it represses gene expression from a tetR- repressible promoter driving YFP, and its mutagenic activity by monitoring mutagenesis at CAN1 in an apnlA strain that has impaired abasic site processing capability. Example 2: Magl-sctetR is a potent mutator ofDNA and avid binder to tetO sites

We chose to use a DNA glycosylase as our mutator enzyme and localize it by fusion with the tet repressor (tetR) which binds the 19 bp tet operator (tetO) sequence. DNA glycosylases normally function as the first step in base excision repair (BER) to remove chemically altered DNA bases. Repair proceeds through excision of the DNA backbone by an apurinic/apyrimidinic (AP) endonuclease. S. cerevisiae has two AP endonucleases, Apnlp and Apn2p. The majority of abasic sites are repaired through Apnlp, and its loss elevates spontaneous mutation rates, as a buildup of unprocessed abasic sites leads to replication fork stalling and recruitment of error-prone translesion polymerases. This faulty repair can lead to both point mutations and frameshifts. We tested two glycosylases that have activity towards undamaged DNA bases: cytosine

DNA glycosylase (CDG), and the yeast 3-methyladenine glycosylase Maglp. CDG is a variant of human UDG that has activity on cytosine in yeast. Maglp is primarily responsible for excising alkylated bases, but has naturally broad substrate specificity and is thought to excise normal base pairs when overexpressed. To test whether these enzymes were active in the W303 strain background, we expressed them from a centromeric plasmid under the control of the galactose-inducible GAL1 promoter and measured mutation rates at the CAN1 locus by fluctuation assay (Fig. 11). We did so in both wild-type (WT) and αρηΙΔ backgrounds, as reduced AP endonuclease activity elevates mutation rates. Maglp increased mutation rates to a much greater extent than CDG in WT (Fig. 6A, 12A). In αρηΙΔ, CDG elevated mutation rates further but Maglp did not due to a severe growth defect in this background (Fig. 6B). At least part of the reason for the reduced potency of CDG relative to Maglp is a difference in relative abundance, as measured by fluorescence of Magi -YFP and CDG- YFP fusions (Fig. 6D, 12B).

Next, we fused both Ma lp and CDG to single-chain (sc) tetR. Normal tetR forms a dimer, so fusion to tetR may result in impaired enzymatic or binding activity due to loss of conformational freedom. Fusing the C-terminal end of a monomelic enzyme to sctetR - a tandem repeat of the tetR connected via a peptide linker- eliminates this problem (see Fig. 2A). Expression of Magl-sctetR but not CDGsctetR elevated mutation rates in an apnlA background (Fig. 3A, 12A). We reasoned cells expressing Magl-sctetR grow in the apnlA because Maglp activity is somewhat compromised by this fusion. Fusion of Maglp per se did not reduce activity because apnlA strains expressing Magl-YFP do not grow (Fig. 6C). Magi -sctetR- YFP fusion had lower abundance as compared to a Magl-YFP fusion (Fig. 6D), but the reduced activity may also be due to lower enzymatic activity.

We introduced a tetR-repressible promoter driving YFP in cells expressing Magl-sctetR to confirm it bound tetO sites. When doxycycline (dox) is added, it binds to and reduces the affinity of sctetR for tetO, relieving the repression and increasing YFP expression. Ma l-sctetR repressed YFP expression to nearly the same extent as sctetR (Fig. 6E). Therefore, Magl-sctetR has both mutagenic and specific DNA binding activity.

Example 3: TaGTEAM elevates the loss of function mutation rate in a 20 kb region surrounding an integrated 240x tetO array To target Magl-sctetR, we integrated a non-recombinogenic 240x tetO array (with each

19 bp tetO site separated by 10-30 bp of random sequence) into the right arm of chromosome I (chrl: 197000). No essential genes are present on the centromere-distal side of this site; on the centromere-proximal side the first essential gene is 21 kb away. To monitor targeted mutations, the K. lactis (Kl)URA3 marker was integrated at various distances surrounding the array (Fig. 2A). To monitor or control the occurrence of chromosomal rearrangements that delete KIURA3 and a large portion of the nearby sequence, the HIS3 marker was placed centromere-distal to the array. Untargeted mutations were monitored at CAN1 on chromosome V. To control for locus and marker- specific effects, we also monitored mutation rates at KIURA3 in the presence of dox and in the absence of the tetO array. Mutation rates at the targeted locus (Fig.2B) were 3.1 x 10^"5 cell^"1 gen^"1 (generation ¹), a

>800-fold increase over the 3.9 x 10^"8 cell^"1 gen^"1 mutation rate measured in the absence of Magl-sctetR. Magi- sctetR expression did not change the mutation rate at CAN1 significantly, but mutation rates at KIURA3 in the absence of the array were elevated 40-fold. This difference indicates Magl-sctetR also causes a locus-dependent increase in the background mutation rate. TaGTEAM creates a region of elevated mutagenesis that spans roughly 10 kb on either side of the array (Fig. 2C). On the centromeric side, mutation rates fall to background at between 17 and 82 kb away. We were unable to probe farther than 15 kb on the telomeric side due to difficulty integrating into the repetitive sub telomeric sequence.

Example 4: TaGTEAM generates both rearrangements and point mutations The loss of function mutation rates measured at KIURA3 do not distinguish between point mutations and rearrangements. To assess the fraction of point mutations at the target locus, we used PCR to probe for the KIURA3 cassette in the genome of mutants (Fig. 2A). A third of mutants at both -8 kb and 0.3 kb were PCR+ (ie. KIURA3 detectable) (Fig. 2C; labels on data points indicate number of PCR+ mutants out of total assayed in parentheses). We sequenced KIURA3 in PCR+ mutants (Table 1). As compared to spontaneous mutagenesis, TaGTEAM generates a broad spectrum of both transitions and transversions. Roughly a quarter of mutants were single base deletions and one complex mutation was observed, containing 3 base substitutions within 10 base pairs.

We found all PCR+ mutants retained the ability to grow on media lacking histidine (his+) while >95% of PCR- mutants were his- and had lost the nearby HIS3 cassette. The correlated loss of the KIURA3 and HIS3 suggests a rearrangement that results in a deletion spanning multiple kbs. We could bias against deletions by selecting for HIS3 using media lacking histidine. Here, mutation rates decreased by roughly one-third in the target region and the fraction of PCR+ mutants increased (Fig. 2C), making point mutagenesis the dominant mutagenic event.

Table 1 sctetR-Fokl +

MAG-sctet sctetR-Fokl + Magi 0.003% MMS Empty vector (N-49) (N=48) (N=70) (N=23) transitions 16.3% 18.8% 31.4% 21.7%

TA>CG 6.1% 14.6% 1.4% 13.0%

CG>TA 10.2% 4.2% 30.0% 8.7% transversions 59.2% 52.1% 64.3% 47.8%

TA>GC 0.0% 6.3% 2.9% 17.4%

GOTA 26.5% 25.0% 40.0% 21.7%

TA>AT 18.4% 16.7% 17.1% 4.3%

GOCG 14.3% 4.2% 4.3% 4.3% deletions 24.5% 29.2% 4.3% 30.4% To further explore the nature of the difference between Ma l-sctetR and sctetR-Fokl, the loss of the nearby HIS3 marker was monitored, the K1URA3 cassette was PCR amplified from the genomes of mutants with K1URA3 present initially either at 0.3 kb or -8 kb (where '-' denotes the centromere-proximal side of the array). This assay distinguishes PMs in K1URA3 from K1URA3 deletion due to GCRs. Among other mechanisms, GCRs could include large scale deletions mediated by repetitive regions introduced during marker integration (Fig. 7) or terminal deletion of the end of chromosome I followed by de novo telomere addition. This assay does not identify events that delete K1URA3 but leave HIS3 intact. A third of Magl-sctetR generated mutants at both locations are PCR+ (ie. K1URA3 is detectable by PCR) and HIS+ (Table 2), consistent with GCRs occurring at a roughly equal rate to PMs. The strong correlation between PCR+ and HIS+ status in mutants indicates HIS+ mutants occur because of PMs in 1URA3 and HIS- mutants are caused by deletion of both K1URA3 and HIS3. To test whether the deletion of both markers was due to an ectopic recombination event involving the short repetitive homologous regions within the integrated array construct (Fig. 7), K1URA3 was introduced in adjacent to an 85x tetO array where no repetitive homology regions were present. This construct had a similar mutation rate but almost all mutants appeared to be PMs, as 11 of 12 mutants are PCR+ (Table 2). Therefore most Magl- sctetR generated mutations are due to either PMs or HR-mediated deletions encompassing K1URA3.

Table 2: Different types of loss of function mutations are determined by genotype and phenotype

Example 5: Mutations are created during repair of targeted damage by homologous recombination Two features of TaGTEAM are inconsistent with the model that Magl-sctetR-mediated point mutations are generated by increased abasic site generation leading to mutagenesis via trans-lesion synthesis (TLS) during replication. First is the long-range point mutagenesis; given Magi is tethered to sctetR by a short (~ 20 nm) peptide linker, it is unclear how it acts to create abasic sites in a 20 kb region flanking the array with roughly equal frequency. Second is the combination of point mutations and rearrangements; TLS of isolated abasic sites should not trigger the large deletions observed. The importance of promoting TaGTEAM' s desirable point mutations motivated us to explore an alternative model whereby Magi damage generates intermediates for HR repair which are then repaired in an error-prone manner. Indeed, when we expressed CFP- tagged Rad52 in exponentially growing Magl-sctetR-expressing cells,we observed fluorescent foci indicative of HR repair whose number increased in an array dependent manner (Fig.3A). Magi generated abasic sites could lead to double-strand ends (DSEs) through fork collapse) or abasic sites clustered within a few helical turns could lead to double strand breaks (DSBs) directly. While the repair of such intermediates by HR is generally error-free, it can be error- prone in certain circumstances. In the presence of repetitive sequence, incorrect homology choice can lead to rearrangements. Repair of DSBs generated by the HO endonuclease has been shown to generate point mutations that are dependent on the error-prone polymerase ζ (REVS- REV? in S. cerevisiae), as has repair of I-Scel induced DSBs in the presence of DNA base pair- damaging agents. We hypothesized that targeted point mutations were occurring through the

HR-dependent localized hypermutagenesis (LHM) process where resectioning of broken ends by EXOl or the SGS1-TOP1- RMI3 complex exposes ssDNA, which is used to search for homology in a iMD52-directed process. Any damage of ssDNA requires lesion bypass by Pol ζ during re-synthesis generating point mutations (Fig. 3B). Since resectioning can proceed many kb from a break, this can explain long-range point mutations. While a DSB is pictured, similar resectioning can occur with a DSE intermediate.

The LHM model predicts that RAD 52, REV 3, and the exonuclease activity of either SGS1 or EXOl are necessary for targeted mutagenesis (Fig. 3B). We measured the mutation rate at -8 kb and 0.3 kb in deletion backgrounds of each repair enzyme with and without selection for HIS3 and subtracted it from the mutation rate in the same deletion background lacking the array (Fig. 3C). This "targeted mutation rate" accounts for global changes due to the deletion. Regardless of selection for HIS3, all targeted mutagenesis requires RAD52, confirming HR as the key repair process. Under selection for HIS3, the targeted point mutants that predominate depend absolutely on REV3 (Pol ζ) and SGS1 + EXOl (resectioning activity).

The RAD52 dependence even without HIS3 selection implies that large deletions spanning both HIS3 and KIURA3 are HR-dependent. When integrating the array and mutator, short repetitive sequence elements were introduced that could explain these correlated deletions (see Fig.7 for a description of how these elements could combine to delete sections of the targeted region). To confirm their involvement, we integrated an 85x tetO array with KIURA3 that lacked the repeated sequences. We found a similar targeted mutation rate (2.4 x 10^"5 cell^"1 gen^"1), but almost all mutants (11/12) were PCR+, suggesting the repeated sequences are responsible for almost all rearrangements. The tetO sites within the array could also cause aberrant recombination, leading to changes in array size or deletion in mutants. Still, PCR+, HIS+ mutants always contain an array as probed by fluorescent foci formed by localized tetR- YFP (Fig.13). On the other hand, sctetR-Fokl damage may not always elicit the checkpoint because the sole signal comes from the DSB, and a single DSB is will not always activate the Mecl/Rad53 pathway [26] . Support was initially sought for this hypothesis by measuring growth rates and DNA content in exponentially growing populations expressing either mutator (Table 3 and Fig. 8)._Without mutator expression, 57% and 43% of cells have 1C and 2C DNA content, respectively. Expression of Magl-sctetR results in an increase of 2C cells to 59% indicating significant checkpoint activation and G2 arrest. Interestingly, sctetR-Fokl causes a slight increase in 2C cells to 47%. This was interpreted as evidence of error-free HR repair of DSBs generated even by sctetR-Fokl.

Table 3: Cell cycle distribution and growth rate of mutator expressing strains mutator strain toxin % 1C / % 2C growth rate +/- SD [hr ¹] none WT none 57/43 0.34 +/- 0.01 sctetR-Fokl WT none 53/47 0.27 +/- 0.02 sctetR-Fokl smll ddc2 none 56/44 0.26 +/- 0.05

Magl-sctetR WT none 41/59 0.28 +/- 0.02

Magl-sctetR smll ddc2 none 55/45 0.28 +/- 0.05 sctetR-Fokl + Maglp WT none 49/51 0.28 +/- 0.04

sctetR-Fokl + Maglp smll ddc2 none 40/60 0.16 +/- 0.04 sctetR-Fokl WT MMS 53/47 0.24 +/- 0.03

sctetR-Fokl WT HU 42/58 0.18 +/- 0.02 sctetR-Fokl WT DTT 53/47 0.24 +/- 0.05

Example 6: Targeted Fokl leads to rearrangements but not point mutations

If the sole role of Magl-sctetR were to create substrates with DNA ends to be repaired by HR, then creating DSBs in the array using an endonuclease might be sufficient for targeted mutagenesis. While site- specific endonucleases have been associated with neighboring damage such enzymes repeatedly cleave the DNA until mutagenic repair of the recognition site prevents further cleavage. Magl-sctetR generates significantly fewer Rad52-CFP foci-containing cells then the site-specific HO endonuclease (Fig. 3B). To better mimic this infrequent damage at the array, we created a C-terminal fusion of the nuclease domain of Fokl to sctetR and expressed it in a strain containing the 240x array and KIURA3 marker at various positions (Fig.5 A). We expected lower efficiency cleavage because the monomeric sctetR-Fokl must dimerize to be active, and potential partners bound to the array may not be optimally spaced. Similar to Magl- sctetR, sctetR-Fokl elevated the mutation rate at the target 620- fold (Fig.5B). However, sctetR- Fokl had no effect on the background mutation rate, either at CAN1 or in the absence of the array. In addition, sctetR-Fokl exhibited an asymmetric distance dependence profile and very few mutants (2/48) were PCR+ (Fig.5C). While the fraction of cells with Rad52-foci in cells experiencing sctetR-Fokl damage at the array was similar to Magl-sctetR (Fig.5D), RAD52 deletion did not completely eliminate targeted mutagenesis in the absence of HIS3 selection (Fig.5E). Therefore, a large fraction of mutations created by sctetR-Fokl are RAD52- independent rearrangements. As expected, this rearrangement did not require the short repetitive sequences present near the array because their elimination did not decrease the mutation rate (2.7 x 10^"5 cell^"1 gen^"1) and most mutants remained rearrangements (2/12 were PCR+). Under HIS3 selection mutation rates throughout the target region were further decreased as compared to Magl-sctetR (Fig. 5C). The remaining mutagenesis was independent of Pol ζ (Fig. 5E) and still predominantly rearrangements (8/32 were PCR+). Therefore, processing of Fokl-generated damage - presumably DSBs - results in loss of KIURA3 function largely through RAD52- independent rearrangements rather than the LHM process in Fig. 3B. Example 7: Checkpoint activation and genome wide DNA damage are sufficient to bias repair towards error-prone HR that generates point mutations

Since sctetR-Fokl damage increases Rad52 foci (Fig. 5D), much of it must be repaired via HR without mutating KIURA3. Understanding why these HR repair events do not lead to point mutations and why the dominant mutagenic event is RAD52 -independent rearrangements could allow us to increase point mutations and potentially use any DSB to generate them. We hypothesized that differences between sctetR-Fokl and Magl-sctetR were either due to the nature of the break intermediate or the cellular context in which the break was repaired. In support of the second hypothesis, Magl-sctetR, but not sctetR-Fokl, has a non-specific DNA damaging activity that increases background mutation rates and increases the fraction of cells with Rad52-CFP foci in the absence of the array. To test if the non-specific DNA damage activity of Magl-sctetR explains the difference in types of mutations generated by each mutator we co-expressed untargeted Maglp with sctetR-Fokl (Fig. 4). Maglp coexpression was sufficient to switch mutations generated by sctetR-Fokl to predominantly point mutations (11/12 were PCR+). HIS3 selection caused no drop in the observed mutation rate, and like Magl- sctetR, targeted mutagenesis was ftEVJ-dependent. The mutation spectrum was also similar to Magl-sctetR (Table 1), consistent with mutations occurring at bases damaged by Magi .

We hypothesized the non-specific DNA damage activity of Maglp promotes HR- mediated point mutations in at least two ways: 1) by activating the G2/M DNA damage checkpoint, biasing repair towards HR and 2) by creating DNA lesions that must be

resynthesized after resection using Pol ζ. Repair pathway choice is affected by checkpoint activation, and LHM surrounding a I-Scel-generated DSB occurs only with the addition of the DNA methylating agent methyl methanesulfonate (MMS).

SYTOX green staining of DNA in growing cells showed more cells with 2C DNA content when Magl-sctetR or Magi (and sctetR-Fokl) was expressed versus sctetR-Fokl alone, consistent with increased G2/M checkpoint activation. We then measured mutation rates in a smll ddc2 background deficient in the G2/M DNA damage checkpoint. Targeted mutagenesis by Magl-sctetR or sctetRFokl and Maglp was completely eliminated under selection for HIS3 (Fig.4B), confirming that point mutagenesis depends on checkpoint activation. In the absence of HIS3 selection (Fig. 4A), mutation rates in strains expressing sctetR-Fokl increased significantly to 5.79 x 10^"4 cell^"1 gen^"1 (- MMS) and 6.51 x 10^"4 cell^"1 gen^"1 (+MMS). Therefore, without checkpoint activation his- rearrangements increase, likely through an HR-independent pathway. Strains co-expressing sctetR-Fokl and Maglp did not show an increase in mutation rate but mutations switched from point mutants (11/12 PCR+) to rearrangements (0/12 PCR+). To see if checkpoint activation was sufficient to shift the mutagenic outcome of a sctetR-Fokl break toward HR-mediated point mutagenesis, we added hydroxyurea (HU) to activate the checkpoint without creating lesions. The addition of HU, which depletes nucleotide pools leading to fork stalling and collapse, to cells expressing sctetR-Fokl (Fig . 4) decreased the mutation rate in the absence of selection 10-fold, such that HIS3 selection no longer had any effect on the mutation rate. Unlike Maglp or Magi- sctetR, checkpoint activation via HU addition decreases HR-independent rearrangements without adding REV3 -dependent point mutations.

Finally, we added MMS to generate DNA lesions in cells experiencing sctetR-Fokl induced breaks at a concentration (0.001%) we found has minimal impact on growth. Mutation rates under HIS3 selection increased in a i?£V3-dependent manner and 6/10 mutants were PCR+, consistent with an increase in the rate of point mutagenesis. Increasing the MMS concentration to 0.003% further increased the mutation rate to 5.14xl0^"5 cell^"1 gen^"1, with 9/12 HIS+ mutants even without HIS3 selection. Further increases in MMS did not affect the mutation rate and growth was impaired. The mutation spectrum generated by MMS (Table 1 ) was different from Magi or Magl-sctetR and consistent with MMS damage occurring at cytosine residues (>70% of base pair substitutions were CG>TA or CG>AT).

The above results indicate Maglp' s ability to generate abasic sites genome-wide is sufficient to switch mutagenic repair of a sctetR-Fokl generated DSB from a Ku-dependent GCR process to a REV3-dependent PM process. While an important function of those abasic sites is to aid in activating the Mecl checkpoint, it remains unclear whether those specific lesions are also required for mutagenesis. The latter would be unnecessary if checkpoint activation upregulates and/or licenses Pol ζ for resynthesis of resected DNA, akin to DinB upregulation and licensing by the SOS and RpoS stress responses, respectively, in E. coli [13, 14] . Similarly, in S. pombe, checkpoint activation enables DinB to associate with the checkpoint clamp, bind chromatin, and perform mutagenic TLS [35]. Alternatively, those lesions might be necessary if they function in much the same way as UV- or MMS-induced lesions in ssDNA, causing PMs by recruiting Pol ζ during re-polymerization of resectioned DNA. Studies have shown a role for Pol ζ in mutagenic recombinational repair of DSBs even in the absence of these agents. To test these two possibilities, genotoxic chemicals that either elicit checkpoint activation or generate ssDNA lesions, but do not do both were utilized. Hydroxyurea (HU) depletes nucleotide pools, leading to replication fork stalling and multiple DSBs that activate the ATR/Meclp checkpoint. MMS is an alkylating agent that damages base pairs and triggers the checkpoint at high but not low doses. Dithiothreitol (DTT) was utilized as a general stress control. DTT is a reducing agent that leads to metabolic stress and growth inhibition without activating the checkpoint or damaging base pairs. Toxin dose was determined at or just below the lowest dose that can slightly alter the growth rate of cells expressing sctetR-Fokl (Table 3). 0.001 % v/v MMS and 0.75 mM DTT had no effect on the cell cycle distribution while 4.5 mg/mL HU caused cells to delay throughout S and in G2 (Table 3 and Fig.8). These low concentrations may correspond to doses commonly experienced by cells in a natural

environment, and allow us to explore synergistic effects of low-level global stress and

(infrequent) introduction of DSBs.

Next, mutation rates were measured in sctetR-Fokl expressing cells grown with these chemicals in wild type, rev3, and ddc2 smll backgrounds to isolate the roles of Pol ζ and checkpoint activation leading to S/G2 delay (Fig.4). With MMS, targeted mutagenesis increases 3-fold under HIS3 selection and depends on REV3 and DDC2 (Fig. 4B). The number of PCR+ mutants increases significantly from 25% (8/32) with sctetR alone to 60% (6/10) denoting a shift to PMs. Because MMS does not increase checkpoint activation, some subset of the error-free HR repair events associated with sctetR-Fokl breaks must now lead to PMs. In contrast, 4.5 mg/mL of HU has no effect on the targeted mutation rate under selection and mutations are independent of REV3. Further, the mutation rate in the absence of selection is similar to the rate under selection, evidence that robust checkpoint activation occurs and prevents Ku-dependent GCR events. Addition of 0.75 mM DTT does not change the repair outcome of sctetR-Fokl generated DSBs (Fig. 4). This is true even at doses that severely effect growth (2.5 mM).

Together, these results imply that PMs require both Mecl checkpoint activation to initiate HR and extensive resectioning and the introduction of DNA lesions that recruit Pol ζ and lead to mutagenic TLS.

As shown in the Examples section above, Applicant has followed the fate of a DSB through two branched decision points that depend on the cellular context in which repair occurs (Fig. 5). Most DSBs are repaired in an 'error-free' way via a canonical pathway of ATR/Meclp checkpoint activation followed by an HR-dependent process. Experiments showed that cell populations generating DSBs at an approximate rate of 0.1 cell^"1 gen^"1 only exhibit a small reduction in growth (Table 3) and otherwise appear phenotypically indistinguishable from a wild type population - a testament to the speed and efficiency of this repair. However, if checkpoint activation is weak, mutagenesis is dominated by Ku-dependent repair through NHEJ or de novo telomere addition (Fig. 5A). If checkpoint activation is eliminated directly (ddc2 smll) or indirectly (sgsl exol), the targeted mutation rate soars to >10^"3 cell^"1 gen^"1, approaching the DSB generation rate and consisting of Ku-dependent GCRs (Fig. 4A). This constitutes the first decision, a switch between checkpoint activation followed by HR and Ku-dependent GCRs. Simply disabling HR (as in a rad52 background) does not cause this switch (Fig. 3B), suggesting a role for checkpoint-dependent but HR and Ku-independent mechanisms of DSB repair [36] that do not mutate K1URA3.

The second decision point is a switch between error-free and error-prone HR repair (Fig. 5B-D). Even at low levels, agents that introduce lesions in resected ssDNA can drive the formation of PMs that occur during Pol ζ-mediated TLS across those lesions. This is the origin of the targeting effect. HR mediated deletions also occur at the target. Such events delete K1URA3 (Fig. 7), or potentially delete of portions of the array mediated by the repeated 19 bp tetO sites. Still, PM events do not lead to significant loss of the array as PCR+ mutants still contain an intact tetO array, as measured by the ability to bind tetR-GFP and generate a bright spot (data not shown). When both the first switch to checkpoint activation and the second switch to error-prone HR occur, the type of mutation caused by DSB repair changes dramatically.

In concert, the two DNA damage stress-dependent switches outlined here (Fig. 5) result in a similar functional outcome as stress-induced mutagenesis pathways identified in E. coli [12]. However, unlike in E. coli, the connection between stress and mutagenesis does not require active signaling and the particular stress must lead to DNA lesions. This is distinct from previously reported instances of stress-induced mutagenesis in yeast [37], which have been associated with mutagenic NHEJ [ 15] . These findings extend beyond previous work connecting mutagenic repair of DSBs with DNA lesions [9] to perhaps more frequently experienced levels of damage that do not significantly inhibit growth, and provide a functional connection between checkpoint activation and mutagenic TLS [38] . From the standpoint of a growing microbial population, under "good" conditions, a DSB that does not trigger the checkpoint will result in a Ku-dependent GCR. Given the compact genome of S. cerevisiae, this will likely be a fatal event, mimicked by HIS3 selection. Its relative infrequency, though (<10 cell^"1 gen^_1-ifor HIS3 loss), does not affect the population' s growth. However, even levels of DNA damage that affect growth negligibly, Pol ζ-dependent PMs increase dramatically, shifting the primary mutagenic event to long range PMs. This is different from E. coli, where multiple types of damage lead to the SOS response and multiple general stresses lead to RpoS activation. Checkpoint activation decreases mutation rates by preventing Ku-dependent GCRs, but paradoxically sets up the possibility for DNA damage stress-induced mutagenesis.

The results highlight additional dangers in exposing human tissue to low levels of agents that create DNA lesions. Regions peripheral to the site of delivery of radio- and

chemotherapeutic agents would also experience low levels and are at increased risk for mutagenic repair of a DSB, whether it also arises due to the therapy or is spontaneous in origin. Moreover, the risk could be further heightened if there are other stresses in the cellular environment to promote robust checkpoint activation. Notably, the initial mutations in cancers have been suggested to be due to PMs in oncogenes that are more likely to be tolerated than GCRs [39] . In fact, recent evidence suggests TLS across resected DNA is a source of PMs in higher eukaryotes, including signatures of this event in multiple sequenced human cancers [40] . The location of PMs in these particular lines suggests that native APOBEC proteins are responsible for the ssDNA lesions. Sequencing of 21 breast cancer genomes provides further evidence of local hypermutagenesis by APOBEC generated lesions clustered to a single DNA strand [41] . Because damage is localized and exhibits strand bias, it is completely consistent with extensive DSB resectioning. Perhaps one reason why HR is downregulated relative to NHEJ in higher eukaryotes is to prevent this type of PM during DSB repair. In mice exposed to chronic irradiation, Ku is upregulated and recombination is decreased, suggesting enhanced NHEJ during damage prevents HR [42] . Damage in more complex genomes with repeated regions may result in viable and potentially oncogenic GCR events, but most will be lethal. One advantage of choosing a DNA mutator enzyme such as Maglp is that it provides, for the first time, a way to introduce clustered abasic sites in a pre-defined genomic location and hence more easily follow their fate. Additionally, Magl-sctetR has the unique properties of eliciting strong checkpoint activation in all cells and creating DNA lesions, thereby channeling mutagenic repair events into HR-dependent PMs. While combining sctetR-Fokl breaks with a DNA damage agent like MMS generally mimics Magl-sctetR, there is some evidence of Magi - specific lesion effects. Untargeted Maglp has greater activity than Magl-sctetR (Fig. 6) but does not cause as great an S/G2 shift in the cell cycle distribution (Fig. 8). Increased checkpoint delay may be a signature of clustered abasic site breaks caused by Magl-sctetR, which can have blocked 3' ends that require SAE2 clipping and/or are less efficient for homology search. These ends could explain both the increase in the HR- mediated deletion rate with Magl-sctetR as compared to co-expression of Maglp and sctetR-Fokl (Fig. 4 & 5) and the reason Magl-sctetR generated breaks appear less efficient for Ku-dependent GCRs (Fig. 4). Further, deletion of REV3 increases the targeted mutation rate in the absence of HIS3 selection in cells experiencing clustered abasic site damage from Magl-sctetR, as opposed to cells experiencing a combination of endonuclease breaks and genome- wide abasic site damage (Fig. 4). Taken together this evidence highlights the importance of studying local effects of various types of clustered DNA damage and indicates clustered abasic site damage may be particularly prone to high rate PMs at long range.

Finally, the ability to generate a specific hypermutable region with negligible effects on growth and no required exogenous agents may be a valuable tool for in vivo targeted mutagenesis in eukaryotes. No technique capable of targeting multiple genes for high rate PMs has yet been demonstrated in yeast.

Example 8: Mutation rates using two different mutator constructs while adding chloroacetaldehyde (CAA)

It is demonstrated herein that chloroacetaldehyde (CAA) is useful for increasing the tendency of point mutations especially for sctetR-Fokl repair. The mutation rates with Ma l- sctetR and sctetR-Fokl upon addition of chloroacetaldehyde (CAA) were examined. The data is shown in Figure 9. Mutation rates were monitored at a KIURA3 locus integrated 0.3 kb on the telomere-proximal side of the tetO array, similar to other figures. The numbers refer to

PCR-i-(total) mutants for the KIURA3 marker. Error bars represent 95% c.i. The mutation spectrum with chloroacetaldehyde (CAA) is shown below in Table 7. Table 7

sctetR- Magl-

Fokl sctetR

+100uM

CAA lOOuM

CAA

Example 9: Illustrates the use of the methods of the invention in commercially relevant systems.

The methods of the invention are useful for many purposes including the production of optimized cell lines for use in commercial production, analysis and research. Two examples of cellular optimization utilizing the mutation scheme of the invention are depicted in Figure 10. Constitutive expression of Magl-sctetR from a strong, commonly used promoter resulted in targeted mutagenesis in a variety of carbon sources in prototrophic lab (S288c) and industrial (Ethanol Red) strains of yeast. FIG. 10A shows targeted mutagenesis on a plasmid at the gain of function marker ade2- 1 that reverts through modification of a stop codon shows that TaGTEAM of plasmid-borne genes is possible in yeast. The significant 3-fold decrease upon addition of doxycycline is less than the decrease observed for loss of function mutagenesis, likely because single point mutations are required to change the stop codon in ade2- 1 , but many tRNA mutations elsewhere in the genome can lead to stop codon suppression. The targeted point mutation rate at ade2-l on a plasmid is similar to the rate at K1URA3 on chromosome I, given the K1URA3 target size of 165bp. Fig. 10B shows constitutive expression of Magl-sctetR allows targeted mutagenesis in a variety of carbon sources in prototrophic lab (S288c) and industrial (Ethanol Red) strains of yeast. The mutator and target genes were inserted without the addition of markers. The numbers shown in Figure 10 refer to PCR+(total) mutants. Error bars represent 95% c.i. Example 10: Nocodazole-induced arrest at G2/M checkpoint can elevate mutation rates in the targeted region and still retain targeting specificity.

G2/M checkpoint arrest further increases the targeted PM rate. A way to increase the PM rate generated by TaGTEAM was hinted at by the plateau in the point mutation rate with increasing ssDNA damage, which suggested resectioning might be limiting the mutation rate. We set out to extend the length of the G2 phase of the cell cycle by arresting cells at the G2/M checkpoint using nocodazole. In addition to providing extra time for resectioning, this would force cells into a context that would favor error-prone HR over other repair outcomes, which might be important in certain applications. Extending the length of G2 in all cells had the added benefit of reducing the difference in growth between mutating cells that might have extended activation of the DNA damage checkpoint and non-mutating cells.

Determining if G2 arrest could be used to increase the PM rate required measuring the mutation rate during a single generation when cells are arrested. This is difficult because there is a lag between changes in the mutational event at DNA and its subsequent effect on the selectable protein product; given a mutation in the DNA, current non-mutant proteins must degrade or be diluted through growth for selection to occur. An estimate of this lag is required to accurately estimate the mutation rate during arrest. It is unclear what the degradation of KlUra3p is in vivo, but an upper bound on the length of the lag assumes no degradation and only dilution due to growth. Cells will reduce the amount of non-mutant protein by half in each generation after a mutation event. We assume that after a loss of function mutation at KLURA3, the dilution of wild type KlUra3p to 1/8^ώ its initial level is sufficient to allow for selection by 5'-FOA. This 3 generation lag also limits the extent to which cultures could be kept free of mutants before arrest by growth in media lacking uracil.

Because of the 3 generation lag between DNA state and protein state, mutants that occur before or after arrest cannot be completely eliminated with selection. If the mutation rate before or after arrest is similar to the rate during arrest, then determining which mutants were generated during arrest is difficult because of the large number of other mutants (Fig. 16). To minimize the generation of mutants before and after arrest, we chose to focus on mutagenesis by sctetR-Fokl. sctetR-Fokl has a 30-fold increase in the targeted mutation rate under selection for HIS3 upon addition of MMS (Fig. 15B), and a >100-fold decrease in the targeted mutation rate (Fig. 5B) upon addition of dox (Fig. 16A). This means that pre-existing mutants could be minimized by growth in media lacking MMS prior to arrest, and that grow out mutations could be minimized by growth in media containing dox after arrest.

Pre-existing mutants are particularly difficult because they affect the mutant frequency of a culture exactly as mutants generated during arrest. Grow-out mutants, on the other hand, appear at later generations and so have a limited effect on the final mutant frequency. Under HIS3 selection, the mutation rate in the absence of MMS is 2 x 10^"6 cell^"1 gen^"1 (Fig 5C).

Assuming deterministic mutant generation during pre-growth in media lacking uracil (accurate because the final number of cells is » 2 x 10⁶), the mutant frequency should be 4 x 10^"6 mutants/cell. Given this frequency we chose to arrest 3000 cells, insuring that only 1/100 cultures contained a pre-existing mutant. Arrest at this small number of cells means a long grow out was required in order to measure OD and plate on selection media. Because of the large decrease in mutation rate upon addition of dox, this long grow out shouldn't be a problem, but in order to control for it some cells were transferred immediately from pre-growth to grow-out media without nocodazole arrest (Fig. 16B, 16C). Comparing the mutation frequency generated from this control in the presence or absence of the tetO array also allowed for the determination of the number of pre-existing mutants.

Because prolonged nocodazole arrest leads to a decrease in cell viability (Bekier et al., 2009), the optimal amount of time in arrest was investigated by determining the number of KIURA3 mutants generated at 2, 4, and 6 hours. To control for off-target mutagenesis and to get a sense for the ratio of PMs to other mutation events, arrest was carried out with and without 0.003% MMS addition, with and without the 240x array, and with and without selection for HIS 3. In order to sort out which mutants were generated during arrest (or were pre-existing) and which mutants were generated during grow out, a cutoff was determined above which mutants likely came from the arrest process. This cut-off was set using the mutation rate during grow

-7 - 1 - 1

out, which was determined to be 2 x 10^" cell^" gen^" by fitting mutant counts in strains without an array and without arrest to the Luria-Delbriick (LD) distribution (Fig. 16B). Experiments with and without dilute MMS and with and without selection were combined so that the cut-off could be applied to all conditions investigated. Using the grow out mutation rate, the probability of observing a large number of mutants in a single culture in a set of replicates was calculated using the LD cumulative distribution, and the cut-off (said large number of mutants) was set such that this probability was less than 5% for a set of 8 replicates. Mutant frequencies higher than the cut-off were assumed to represent mutations that occurred during arrest, with multiples of the cutoff corresponding to multiple arrest mutations. The cut-off was roughly half the mutant frequency expected from a mutant generated during arrest. This difference could reflect plating efficiencies or growth defects of mutants as compared to non-mutants during grow-out. Arrest mutations should be Poisson-distributed, which was easily verified by comparing the mean to the variance, and so the distribution of arrest mutations in 8 replicate cultures was fit to a Poisson distribution using MLE to generate a mutation rate during arrest and 95% confidence intervals on that rate (Fig. 17). 4 hour nocodazole arrest increased the mutation rate to 6 x 10^"4 cell^"1 gen^"1 with MMS under selection for HIS3, a 15-fold increase over the rate in the absence of arrest (Fig 15). The combination of MMS and arrest also significantly increased the mutation rate in a strain lacking the array, to a rate that is catastrophic if it were the actual background genomic mutation rate (Daee et al., 2009). Therefore, this increased rate in the absence of the array must be locus-specific and perhaps related to the locus-specific background effect observed with Magl-sctetR expression.

Cells from the "0 hrs" control were transferred directly from pre-growth to grow-out media (Fig 16C), so the cultures that contained a number of mutants greater than the cut-off must represent pre-existing mutants. We saw 1 in 10 cultures containing large numbers of mutants in experiments with the array and without MMS under HIS3 selection. This suggests pre-existing mutants are more prevalent than expected, possibly because the 3 generation phenotypic lag is inaccurate and older mutants are viable during pre-growth. It is also possible that these mutants are unlikely grow out events or even contamination from FOA resistant strains. Whatever their source these false positive events define the lower detection limit of the arrest mutation rate using this assay.

PM rates appear to peak at 4 hours. After this time, mutation rates decreased in the presence of selection and increased in its absence, consistent with the conversion of mutants from His^"1" to His^" status during prolonged arrest. Prolonged nocodazole arrest is associated with decreased cell viability (Bekier et al., 2009). It may be that cells carrying out targeted mutagenesis that partially succeed in generating a mitotic spindle pole and initial chromosome partitioning are more likely to undergo chromosomal rearrangements leading to loss of the HIS3 marker.

In order to better understand the nature of the mutation rate increase caused by nocodazole arrest, the fraction of cells containing Rad52-CFP foci was determined at 2, 4 and 6 hours in the absence of MMS (Fig 18). The fraction of cells containing Rad52-CFP foci roughly triples during arrest, and the majority of these foci are in G2 cells. Interestingly, the fraction of cells in G2/M also triples, possibly explaining the increase in Rad52-CFP foci. This, coupled with the fact that the number of foci does not increase over time in arrest, suggests that arrest does not increase the chance that a G2 cell undergoes a TaGTEAM-induced break. Instead, arrest only increases the fraction of G2 cells. This is in contrast to the increasing fraction of cells (both G2 and M) in the absence of the array that acquire breaks as arrest continues. This fraction of TaGTEAM indepdent foci is small compared to the TaGTEAM induced fraction, but it suggests that prolonged arrest leads to a TaGTEAM-independent increase in HR induction, consistent with prolonged arrest being detrimental to cell viability.

References

1. Georgakilas AG. (2008) Processing of DNA damage clusters in human cells: Current status of

knowledge. Mol BioSyst 4(1): 30-35.

2. Eccles U, Lomax ME, O'Neill P. (2010) Hierarchy of lesion processing governs the repair, double- strand break formation and mutability of three-lesion clustered DNA damage. Nucleic Acids Research 38(4): 1123-1134.

3. Eccles U, O'Neill P, Lomax ME. (2010) Delayed repair of radiation induced clustered DNA damage:

Friend or foe? Mutat Res : Fundam Mol Mech Mutagen In Press, Corrected Proof.

4. Paques F, Haber JE. (1999) Multiple pathways of recombination induced by double-strand breaks in saccharomyces cerevisiae. Microbiol Mol Biol Rev 63(2): 349-404.

5. Heyer W, Ehmsen KT, Liu J. (2010) Regulation of homologous recombination in eukaryotes. Annu Rev Genet 44(1): 113-139.

6. Pfeiffer P, Goedecke W, Kuhfittig-Kulle S, Obe G. (2004) Pathways of DNA double-strand break repair and their impact on the prevention and formation of chromosomal aberrations. Cytogenet

Genome Res 104(1-4): 7-13.

7. Holbeck SL, Strathern J N. (1997) A role for REV3 in mutagenesis during double-strand break repair in saccharomyces cerevisiae. Genetics 147(3): 1017-1024.

8. Yang Y, Sterling J, Storici F, Resnick MA, Gordenin DA. (2008) Hypermutability of damaged single- strand DNA formed at double-strand breaks and uncapped telomeres in yeast saccharomyces cerevisiae. PLoS Genet 4(11): el000264.

9. Yang Y, Gordenin DA, Resnick MA. (2010) A single-strand specific lesion drives MMS-induced hypermutability at a double-strand break in yeast. DNA Repair 9(8): 914-921.

10. Hicks WM, Kim M, Haber JE. (2010) Increased mutagenesis and unique mutation signature

associated with mitotic gene conversion. Science 329(5987): 82-85. 11. Deem A, Keszthelyi A, Blackgrove T, Vayl A, Coffey B, et al. (2011) Break-induced replication is highly inaccurate. PLoS Biol 9(2): el000594.

12. Ponder RG, Fonville NC, Rosenberg SM. (2005) A switch from high-fidelity to error-prone DNA double-strand break repair underlies stress-induced mutation. Mol Cell 19(6): 791-804.

13. Shee C, Gibson JL, Darrow MC, Gonzalez C, Rosen berg SM. (2011) Impact of a stress-inducible switch to mutagenic repair of DNA breaks on mutation in escherichia coli. . Proc Natl Acad Sci U S A 108(33): 13659-13664.

14. Galhardo RS, Hastings PJ, Rosenberg SM. (2007) Mutation as a stress response and the regulation of evolvability. Crit Rev Biochem Mol Biol 42(5): 399-435.

15. Heidenreich E, Novotny R, Kneidinger B, Holzmann V, Wintersberger U. (2003) Non-homologous end joining as an important mutagenic process in cell cycle-arrested cells. EMBO J 22(9): 2274- 2283.

16. Heidenreich E. (2007) Adaptive mutation in saccharomyces cerevisiae. Crit Rev Biochem Mol Biol 42(4): 285-311.

17. Westmoreland J, Ma W, Yan Y, Van Hulle K, Malkova A, et al. (2009) RAD50 is required for efficient initiation of resection and recombinational repair at random, gamma-induced double-strand break ends. PLoS Genet 5(9): el000656.

18. Glassner BJ, Rasmussen U, Najarian MT, Posnick LM, Samson LD. (1998) Generation of a strong mutator phenotype in yeast by imbalanced base excision repair. Proc Natl Acad Sci U S A 95(17): 9997-10002.

19. Boiteux S, Guillet M. (2004) Abasic sites in DNA: Repair and biological consequences in

saccharomyces cerevisiae. DNA Repair (Amst) 3(1): 1-12.

20. Bitinaite J, Wah DA, Aggarwal AK, Schildkraut I. (1998) Fokl dimerization is required for DNA

cleavage. Proc Natl Acad Sci U S A 95(18): 10570-10575.

21. Krueger C, Berens C, Schmidt A, Schnappinger D, Hillen W. (2003) Single-chain tet transregulators.

Nucleic Acids Res 31(12): 3050-3056.

22. Murphy KF, Balazsi G, Collins JJ. (2007) Combinatorial promoter design for engineering noisy gene expression. Proceedings of the National Academy of Sciences 104(31): 12726-12731.

23. Gibbs PE, Lawrence CW. (1995) Novel mutagenic properties of abasic sites in saccharomyces cerevisiae. J Mol Biol 251(2): 229-236.

24. Lau IF, Filipe SR, Soballe B, Okstad OA, Barre FX, et al. (2003) Spatial and temporal organization of replicating escherichia coli chromosomes. Mol Microbiol 49(3): 731-743.

25. Lisby M, Rothstein R, Mortensen UH. (2001) Rad52 forms DNA repair and recombination centers during S phase. Proceedings of the National Academy of Sciences 98(15): 8276-8282.

26. Pellicioli A, Lee SE, Lucca C, Foiani M, Haber J E. (2001) Regulation of saccharomyces Rad53

checkpoint kinase during adaptation from DNA damage-induced G2/M arrest. Mol Cell 7(2): 293- 300. 27. Pomerening JR, Kim SY, Ferrell JJames E. (2005/8/26) Systems-level dissection of the cell-cycle oscillator: Bypassing positive feedback produces damped oscillations. Cell 122(4): 565-578.

28. Pennaneach V, Putnam CD, Kolodner RD. (2006) Chromosome healing by de novo telomere

addition in saccharomyces cerevisiae. Mol Microbiol 59(5): 1357-1368.

29. Chung WH, Zhu Z, Papusha A, Malkova A, Ira G. (2010) Defective resection at DNA double-strand breaks leads to de novo telomere formation and enhances gene targeting. PLoS Genet 6(5):

el000948.

30. Hanna M, Ball LG, Tong AH, Boone C, Xiao W. (2007) Pol32 is required for ροΐζ-dependent

translesion synthesis and prevents double-strand breaks at the replication fork. Mutation

Research/Fundamental and Molecular Mechanisms of Mutagenesis 625(1-2): 164-176.

31. Lydeard JR, Lipkin-Moore Z, Jain S, Eapen VV, Haber JE. (2010) Sgsl and exol redundantly inhibit break-induced replication and de novo telomere addition at broken chromosome ends. PLoS Genet 6(5): el000973.

32. Rattray AJ, Shafer BK, McGill CB, Strathern JN. (2002) The roles of REV3 and RAD57 in double- strand-break-repair-induced mutagenesis of saccharomyces cerevisiae. Genetics 162(3): 1063-

1077.

33. Symington LS, Gautier J. (2011) Double-strand break end resection and repair pathway choice.

Annu Rev Genet 45: 247-271.

34. Paciotti V, Clerici M, Lucchini G, Longhese MP. (2000) The checkpoint protein Ddc2, functionally related to S. pombe Rad26, interacts with Mecl and is regulated by Mecl-dependent

phosphorylation in budding yeast. Genes Dev 14(16): 2046-2059.

35. Kai M, Wang TS. (2003) Checkpoint activation regulates mutagenic translesion synthesis. Genes Dev 17(1): 64-76.

36. Ma JL, Kim EM, Haber JE, Lee SE. (2003) Yeast Mrell and Radl proteins define a ku-independent mechanism to repair double-strand breaks lacking overlapping end sequences. Mol Cell Biol 23(23):

8820-8828.

37. Steele DF, Jinks-Robertson S. (1992) An examination of adaptive reversion in saccharomyces

cerevisiae. Genetics 132(1): 9-21.

38. Koren A. (2007) The role of the DNA damage checkpoint in regulation of translesion DNA synthesis.

Mutagenesis 22(3): 155-160.

39. Loeb LA, Loeb KR, Anderson JP. (2003) Multiple mutations and cancer. Proc Natl Acad Sci U S A 100(3): 776-781.

40. Roberts S, Sterling J, Thompson C, Harris S, Mav D, et al. (2012) Clustered mutations in yeast and in human cancers can arise from damaged long single-strand DNA regions. Mol Cell 46(4): 424-435. 41. Nik-Zainal S, Alexandrov LB, Wedge DC, Van Loo P, Greenman CD, et al. (2012) Mutational

processes molding the genomes of 21 breast cancers. Cell 149(5): 979-993. 42. Kovalchuk O, Hendricks CA, Cassie S, Engelward AJ, Engelward BP. (2004) In vivo recombination after chronic damage exposure falls to below spontaneous levels in "recombomice". Mol Cancer Res 2(10): 567-573.

43. Wang HH, Isaacs FJ, Carr PA, Sun ZZ, Xu G, et al. (2009) Programming cells by multiplex genome engineering and accelerated evolution. Nature 460(7257): 894-898.

44. Esvelt KM, Carlson JC, Liu DR. (2011) A system for the continuous directed evolution of

biomolecules. Nature 472(7344): 499-503.

45. Foster PL. (2006) Methods for determining spontaneous mutation rates. Methods Enzymol 409:

195-213.

46. Lang Gl, Murray AW. (2008) Estimating the per-base-pair mutation rate in the yeast saccharomyces cerevisiae. Genetics 178(1): 67-82.

47. Sarkar S, Ma WT, Sandri GH. (1992) On fluctuation analysis: A new, simple and efficient method for computing the expected number of mutants. Genetica 85(2): 173-179.

48. Stewart FM. (1994) Fluctuation tests: How reliable are the estimates of mutation rates? Genetics 137(4): 1139-1146.

49. Haase SB, Reed SI. (2002) Improved flow cytometric analysis of the budding yeast cell cycle. Cell Cycle 1(2): 132-136.

The foregoing written specification is considered to be sufficient to enable one skilled in the art to practice the invention. The present invention is not to be limited in scope by examples provided, since the examples are intended as a single illustration of one aspect of the invention and other functionally equivalent embodiments are within the scope of the invention. Various modifications of the invention in addition to those shown and described herein will become apparent to those skilled in the art from the foregoing description and fall within the scope of the appended claims. The advantages and objects of the invention are not necessarily encompassed by each embodiment of the invention.

What is claimed is:

Claims

1. A method for in vivo targeted mutagenesis, comprising, selectively introducing localized DNA damage in a preselected region of an organism' s DNA in vivo, biasing repair of the localized DNA damage by targeting a pathway requiring long-range resectioning of the localized DNA damage, wherein the DNA forms a single stranded region during the biasing repair, and selectively mutating the single stranded region to cause targeted mutagenesis, optionally wherein the organism is a eukaryotic organism.

2. The method of claim 1, wherein the localized DNA damage is a double stranded break (DSB).

3. The method of claim 2, wherein the DSB is introduced by a DNA mutator enzyme domain.

4. The method of claim 2, wherein biasing repair of the DSB involves contacting the cell with a compound that elicits DNA damage checkpoint activation.

5. The method of claim 4, wherein the compound that elicits DNA damage checkpoint activation is a chemical checkpoint activator.

6. The method of claim 5, wherein the chemical checkpoint activator is MMS.

7. The method of claim 4, wherein the compound that elicits DNA damage checkpoint activation is an enzymatic checkpoint activator.

8. The method of claim 7, wherein the enzymatic checkpoint activator is Magi.

9. A method comprising, contacting a cell with a fusion protein, wherein the fusion protein is an array specific DNA binding domain and a DNA mutator enzyme domain, wherein the cell has an integrated array of DNA binding sites to which the DNA binding domain is capable of binding.

10. The method of claim 9, wherein the DNA mutator enzyme is a DNA glycosylase.

11. The method of claim 10, wherein the DNA glycosylase is 3-methyladenine glycosylase Ma lp (Maglp).

12. The method of claim 9, wherein the DNA mutator enzyme is a DNA nuclease.

13. The method of claim 12, wherein the DNA nuclease is Fokl.

14. The method of any one of claims 9-13, further comprising contacting the cell with a compound that elicits DNA damage checkpoint activation.

15. The method of claim 14, wherein the compound that elicits DNA damage checkpoint activation is a chemical checkpoint activator.

16. The method of claim 15, wherein the chemical checkpoint activator is MMS.

17. The method of claim 15, wherein the chemical checkpoint activator is Nocodazole, HU or methotrexate.

18. The method of claim 14, wherein the compound that elicits DNA damage checkpoint activation is an enzymatic checkpoint activator.

19. The method of claim 18, wherein the enzymatic checkpoint activator is Magi.

20. The method of claim 9, wherein the array specific DNA binding domain is tetR.

21. The method of claim 9, wherein the array specific DNA binding domain is (sc)tetR.

22. The method of claim 9, wherein the integrated array of DNA binding sites is a tetO array.

23. The method of claim 22, wherein the tetO array is an 85x or 240x tetO array.

24. The method of claim 9, wherein the cell is a yeast cell.

25. The method of claim 9, wherein the cell is a mammalian cell.

26. The method of claim 9, wherein the cell is a plant cell.

27. The method of claim 9, further comprising downregulating non-HR processes in the cell.

28. The method of claim 27, wherein the downregulation of non-HR processes in the cell involves contacting the cell with an inhibitory nucleic acid, wherein the genetic target is optionally Ku70/Ku80 or DNAPkcs.

29. A method comprising, contacting a cell with a fusion protein, wherein the fusion protein is a DNA binding domain and a DNA nuclease domain in order to produce a localized DNA damage, and biasing repair of the localized DNA damage by targeting a pathway requiring long-range resectioning of the localized DNA damage.

30. The method of claim 29, wherein the localized DNA damage is a double stranded break (DSB).

31. The method of claim 30, wherein biasing repair of the DSB involves contacting the cell with a compound that elicits DNA damage checkpoint activation.

32. The method of claim 31, wherein the compound that elicits DNA damage checkpoint activation is a chemical checkpoint activator.

33. The method of claim 32, wherein the chemical checkpoint activator is MMS.

34. The method of claim 31, wherein the compound that elicits DNA damage checkpoint activation is an enzymatic checkpoint activator.

35. The method of claim 34, wherein the enzymatic checkpoint activator is Magi.

36. The method of claim 35, wherein the cell is contacted with a recombinant nucleic acid capable of expressing Magi.

37. The method of claim 30, wherein biasing repair of the DSB involves contacting the cell with an inhibitory nucleic acid, wherein the genetic target is optionally Ku70/Ku80 or DNAPkcs.

38. The method of claim 29, wherein the DNA binding domain is a TAL binding domain or a zinc finger binding domain.

39. The method of claim 29, wherein the DNA binding domain is Cas9 or dCas9.

40. The method of claim 29, wherein the DNA binding domain is fused to an endonuclease and is a TALEN or is multiple TALENs.

41. The method of claim 29, wherein the DNA binding domain is fused to an endonuclease and is Cas9 or dCas9.

42. The method of claim 29, wherein the cell is a yeast cell.

43. The method of claim 29, wherein the cell is a mammalian cell, and optionally is a mammalian stem cell.

44. The method of claim 29, wherein the cell is a plant cell.

45. The method of claim 29, wherein the cell is not a cell of human B lymphocyte lineage.

46. A fusion protein, comprising an array specific DNA binding domain and a DNA mutator enzyme domain, wherein the

DNA mutator enzyme is not a deaminase protein.

47. The fusion protein of claim 46, wherein the DNA binding domain is a TAL binding domain or a zinc finger binding domain.

48. The fusion protein of claim 46, wherein the DNA binding domain is a Cas9 or dCas9 binding domain.

49. The fusion protein of claim 46, wherein the DNA mutator enzyme is a DNA glycosylase.

50. The fusion protein of claim 46, wherein the DNA glycosylase is 3-methyladenine glycosylase Maglp (Maglp).

51. The fusion protein of claim 50, wherein the DNA glycosylase is a uracil DNA glycosylase.

52. The fusion protein of claim 46, wherein the DNA mutator enzyme is a DNA nuclease.

53. The fusion protein of claim 52, wherein the DNA nuclease is Fokl.

54. The fusion protein of claim 52, wherein the DNA nuclease is scFokl.

55. The fusion protein of claim 46, wherein the fusion protein is scFokl-TALe.

56. The fusion protein of claim 46, wherein the fusion protein is sctetR-Fokl.

57. The fusion protein of claim 46, wherein the fusion protein is Magl-sctetR.

58. The fusion protein of claim 46, wherein the fusion protein is Magl-Cas9 or Magl- dCas9.

59. An isolated nucleic acid, comprising a nucleic acid sequence that encodes the fusion protein of any of claims 46-58.

60. An expression vector, comprising the isolated nucleic acid of claim 59.

61. A host cell, comprising the expression vector of claim 60.

62. A plant comprising a mutated germline, wherein the plant is produced according to a method of any of claims 1-45.

63. A human stem cell comprising a mutated germline, wherein the human stem cell is produced according to a method of any of claims 1-45.

64. A method for in vivo targeted mutagenesis in a cell, comprising: delivering a plasmid expressing a Cas9 nuclease and a synthetic single-guide RNA (gRNA) to a target gene of a cell and contacting the cell with a mutator.

65. The method of claim 64, wherein the mutator is methyl methanesulfonate (MMS).

66. The method of claim 64, wherein the plasmid expressing the Cas9 nuclease also includes an expression cassette for expressing the synthetic gRNA.

67. The method of claim 64, wherein the synthetic gRNA is delivered to the cell in a plasmid expressing the gRNA, and wherein the plasmid expressing the gRNA is a separate plasmid from the plasmid expressing the Cas9 nuclease.

68. The method of claim 64, wherein the synthetic gRNA has a CRISPR RNA (crRNA) and a trans-activating crRNA (tracrRNA).

69. The method of claim 64, wherein the Cas9 nuclease is a codon- optimized Cas9 nuclease.

70. The method of any one of claims 64-69, wherein the plasmid expressing the Cas9 nuclease also expresses a DNA mutator enzyme.

71. The method of claim 70, wherein the Cas9 nuclease and the DNA mutator enzyme are expressed as a fusion protein.

72. The method of claim 71, wherein the DNA mutator enzyme is Magi.