- Email: [email protected]
Mutagenesis Assays in Yeast
METHODS 22, 116 –119 (2000) doi:10.1006/meth.2000.1051, available online at http://www.idealibrary.com on
Mutagenesis Assays in Yeast Gray F. Crouse Department of Biology, Emory University, Atlanta, Georgia 30322
Analyzing mutation spectra is a very powerful method to determine the effects of various types of DNA damage and to understand the workings of various DNA repair pathways. However, compiling sequence-specific mutation spectra is laborious; even with modern sequencing technology, it is rare to obtain spectra with more than several hundred data points. Two assay systems are described for yeast, one for insertion/deletion mutations and one for base substitution mutations, that allow determination of specific mutations without the necessity of DNA sequencing. The assay for insertion/deletion mutations uses a variety of different simple repeats placed in frame with URA3 such that insertions or deletions lead to a selectable Ura ⫺ phenotype; essentially all such mutations are in the simple repeat sequence. The assay for base substitution mutations uses a series of six strains with different mutations in one essential codon of the CYC1 gene. Because only true reversions lead to a selectable phenotype, the bases mutated in any reversion event are known. The advantage of these assays is that they can quantitatively determine over several orders of magnitude the types of mutations that occur under a given set of conditions, without DNA sequencing. © 2000 Academic Press
Powerful genetics, ease of manipulation, and availability of the complete nucleotide sequence make Saccharomyces cerevisiae an excellent organism for the study of DNA damage and repair. One of the best methods for initial characterization of genes involved in DNA damage and repair is to analyze the spectrum of mutations created in their absence. Similarly, the mutagenic effects of external agents can be measured by analyzing their mutation spectra. A number of different systems in yeast have been used to assay mutations. One of the simplest and most commonly used assays is selection for inactivation of the CAN1 gene by growth on canavanine (1). The tacit assumption has been that a majority of inactivating mutations in the CAN1 gene would normally be due to base substitution errors, although recent sequencing of can1 mutations has revealed that in the absence of mismatch repair, a 116
majority of the mutations are frameshift mutations (2). The SUP4-o gene has been used as a mutational target in many studies from the Kunz laboratory (3, 4), and the URA3 gene has been used as a target in several laboratories (3, 5). Frameshift mutations have been studied by selection for reversion of the frameshift mutations hom3–10 (6) and lys2⌬Bgl (7). The difficulty with all of these assays is that to determine a mutational spectrum it is necessary to sequence each mutant gene. Even with modern sequencing methods, the number of events that can be analyzed is quite limited. Measurement of forward mutation rates in a gene generally requires that the entire gene be sequenced, which is a considerable task for the CAN1 gene, approximately 1.8 kb in size, but much easier for the small SUP4-o gene. The frameshift mutations mentioned above appear to revert mainly via compensating internal frameshift mutations which are limited to a region surrounding the original mutation that will restore the open reading frame and not interfere with function. Reversion of the lys2⌬Bgl mutation has been extensively studied and most reversion events occur within a 150-bp window (8); there is also a relatively short reversion window for the hom3–10 mutation, although the exact length over which compensating frameshift mutations can occur has not been as extensively explored (2). The article describes the use of two assays in yeast that can examine both frameshift (insertion/deletion) mutations and base pair substitutions with great specificity without sequencing. Most of the insertion/ deletion mutations observed in wild-type strains occur in repeated stretches of DNA and are apparently due to slipped mispairing (3, 8). Thus an assay designed to examine slipped mispairing mutations can serve as an assay for the general occurrence of frameshift or insertion/deletion mutations. A convenient assay for detecting simple repeat instability has been developed in the Petes laboratory (9 –12). This assay depends on a slippage event in a tester region either putting Escherichia coli lacZ in-frame and thus forming blue colo1046-2023/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
MUTAGENESIS ASSAYS IN YEAST
nies or shifting a URA3 gene out of frame, resulting in colonies resistant to 5-fluoroorotic acid (5FOA). The tester region can range from mononucleotide repeats to much longer repeats, but cannot be a multiple of 3. The assay with the URA3 gene is more convenient, since only cells with a slippage event will survive, and this assay has been used to examine the effects of mismatch repair, transcription, repeat length, and repeat size on insertion/deletion mutations (10 –14). A plating assay reveals the colonies resulting from a slippage event, but does not directly give information about the type of slippage event. However, a simple polymerase chain reaction (PCR) can determine the length of the tester region, and this information is sufficient to reveal the type of slippage event that occurred (10). A slippage assay has also been developed by placing various mononucleotide runs in the LYS2 gene, with most of the LYS2 mutations occurring in the added sequences (15). This assay system shares a number of characteristics with the one described here, but does not yet have a variety of substrates available. Determination of base pair substitution mutations has traditionally relied on sequencing of mutant genes obtained in a forward mutation assay, and unlike the frameshift reversion assays, it is necessary to sequence the entire gene for each individual mutation. However, Hampsey developed a tester system for yeast that can detect all base pair substitutions with a simple plating assay, and this assay system makes it possible to develop mutation spectra for base pair substitutions without sequencing (16). This assay uses six strains with a different base pair mutation in either position 1 or 2 of the essential codon Cys-22 of the CYC1 gene. Only a true reversion will enable yeast to use glycerol and thus a selective plating on glycerol plates will reveal the reversion rate for a specific base pair mutation.
MEASUREMENT OF INSERTION/DELETION MUTATION RATES All tester plasmids are derivatives of the pSH44 plasmid (9). An extensive series of test sequences differing in composition, length, and length of repeat in this vector have been made in the Petes laboratory (9, 11, 12, 17). The advantage of having the test sequence on a plasmid is that the plasmid can then be easily moved into different genetic backgrounds. A wide variety of genetic backgrounds can be used, but the system imposes certain constraints. The plasmid contains a TRP1 marker and selection is for a Ura ⫺ phenotype. In addition, the URA3 gene on the plasmid is under control of a LEU2 promoter, which requires growth in the absence of leucine and threonine for derepression of the promoter. Therefore the host strain must be trp1 ura3 Leu ⫹ Thr ⫹.
117
As with any mutation rate measurement, it is necessary to determine the number of mutants in many separate cultures. Instability of the test sequence in a plasmid is high enough in most cases that measurements can be made directly from colonies grown on plates without having to grow a liquid culture (12), thus decreasing the amount of effort involved. With stable test sequences, it is necessary to grow independent cultures in liquid to obtain enough cells for measurement. The procedure for mutation determination is as follows (12): Cells containing the test plasmid are plated on solid medium lacking only tryptophan. Preparation of medium is by standard procedures (18). After 3 days of growth at 30°C entire colonies are suspended in water and appropriate dilutions are plated on solid medium containing 0.1% 5FOA and lacking leucine, threonine, and tryptophan; viable cells are determined by plating dilutions on the same medium lacking 5FOA. Colonies are counted after incubation at 30°C for 3 days. Mutation rates are then determined using the method of the median (19). Because the test region is almost always mutated by insertion or deletion of various numbers of repeat units or by a mutation in the URA3 gene, the type of mutation can be revealed by determining the length of the altered region. Changes in length reflect a slippage event, and mutants without a change in length have usually undergone a mutation in the URA3 gene, which can be verified by sequencing (11). Length determination is most easily done by performing PCR on colonies or isolated genomic DNA using the primers 5⬘-CCAATAGGTGGTTAGCAATCG and 5⬘-GTTTTCCCAGTCACGAC and measuring the size of the resulting PCR product on a DNA sequencing gel (12). If necessary the PCR product can be sequenced directly. The advantage of this assay for determination of insertion/deletion mutation rates is that the specificity of insertion/deletion mutation can be determined without sequencing. For example, for a mononucleotide repeat, the ratio of rate of mutation in an msh2 strain to that in an msh3 strain is 48, whereas the same ratio for a tetranucleotide repeat is 1.2 (11). The ratio of deletions to insertions for the tetranucleotide repeat in an msh2 strain is 2.0, whereas the ratio for an msh3 strain is 3.5 (11). Thus from this assay one learns that MSH3 is relatively more active on a tetranucleotide loop than on a single-base-pair loop and that there is a difference in the action of MSH2 and MSH3 on a tetranucleotide loop. New target sequences can also be easily generated. For example, the mononucleotide repeat that has been used is a mononucleotide GC repeat (11), but the other three possible mononucleotide sequences could be used as well, which would test the effect not only on AT pairs, but on the strandedness of the sequences. One disadvantage of this form of the assay is that mutations in the URA3 gene will also lead to 5FOA
118
GRAY F. CROUSE
resistance. Resistance due to mutation is seen in as many as half of the 5FOA-resistant colonies from msh6 strains containing repeats inefficiently recognized by MSH6. Mutations in the URA3 gene are easily recognized as the most likely cause of 5FOA-resistant strains that do not show a change in repeat length. If necessary, such background can be eliminated by the use of an out-of-frame lacZ gene that is put in-frame by a change in repeat length (9). It should also be noted that although the assay is much easier to use when the test region is contained on a plasmid, the mutation rate is much lower when the sequence is in the chromosome and it is not clear that the mutational spectra are the same in the two states (10). Therefore in some cases, it may be desirable to put the test sequence in the chromosome (9, 13, 14).
MEASUREMENT OF BASE PAIR SUBSTITUTION RATES The function of the yeast CYC1 gene is absolutely dependent on having a cysteine residue at amino acid 22 of the protein. Strains mutant at Cys-22 will not grow on nonfermentable carbon sources and reversion of such strains can be selected by growth on glycerol. All revertants must restore the Cys at position 22; therefore for a given mutant strain, the mutation that restores function is known (16). A set of six strains that can together detect all possible base substitutions is available in an isogenic background (16). There are several important aspects of the genetic background of this tester system. The assay strains need to be diploid. It appears that if haploid strains are used, various recessive mutations can accumulate and lead to growth phenotypes that give irreproducible results (16). The diploid cells are made from a haploid cell containing the Cys-22 mutation and from a common haploid of opposite mating type. Both haploids have to be deleted for the related iso-2-cytochrome c gene (CYC7), or otherwise overexpression of CYC7 or recombination between CYC1 and CYC7 could lead to a Cyc ⫹ phenotype. In addition, the haploid cell of opposite mating type must be deleted for the CYC1 gene, resulting in diploids that are hemizygous for the Cys-22 mutant genes. All of the diploids are ura3, allowing the URA3 gene to be used as a selectable marker for gene disruptions, and the haploids of opposite mating type have either a leu2 or his1 marker, allowing easy selection for diploids. The assay procedure is as follows. The haplid strains containing the Cys-22 mutation are YMH2-7 (MAT␣ cyc1-22 ura3-52 leu2-3 112cyh2 cyc7-67), with the only difference between the six strains being the exact mutation in Cys-22, and B-7462 (MATa cyc1-1 cyc7-67 ura3-52 his1-1 can1-100). The cyc1-1 and cyc7-67 mutations are complete deletions. Any desired genetic
modifications are made on the haploid strains and then one of the YMH strains is mated with B-7462 by mixing the two strains, incubating on YPD plates (18) at 30°C overnight, and then selecting diploids by replica plating on SC minimal plates (18) supplemented with uracil. After purification of the diploids, all strains are grown for 2 days in YPD medium at 30°C to approximately 7 ⫻ 10 7 cells/ml. Five-milliliter cultures are harvested, washed with sterile doubly distilled H 2O, and resuspended to 1 ml in doubly distilled H 2O. A portion of each culture is diluted in doubly distilled H 2O and plated on two YPD plates to determine the number of cells in the culture. For most cultures, the remainder of the culture is plated onto YPGD plates, containing 3% (v/v) glycerol and 0.1% (w/v) glucose (16), at a density of no more than 9 ⫻ 10 3 cells/mm 2. For some cultures with an increased reversion rate, a tenfold dilution of the original culture is plated onto two YPGD plates at a density of no more than 7 ⫻ 10 2 cells/mm 2. YPD plates are counted after 2 days at 30°C and YPGD plates are counted after 7 days at 30°C. If cells are to be treated with mutagens, they are harvested and treated just before selective plating (16). Reversion rates are determined using a Luria and Delbru¨ck fluctuation analysis (20). The reversion rate of the unmodified diploid strains is very low, ranging between 10 ⫺10 and 10 ⫺9. To measure rates this low, it is necessary to plate all of the cells in each culture, and then use the number of cultures giving no colonies to determine the rate (20). A 5-ml culture can be plated on a total of three 150-mm Petri plates, without having problems with excessive density of cells. The advantage of this assay, in addition to eliminating the need for sequencing, is that very large differences in effects can be measured. For example, using the diploid strain YMH54, containing a GC substitution for a TA pair, it was found that deleting the mismatch repair gene MSH2 increased the rate of GC 3 TA transversions more than 2000 fold, whereas the MSH3 gene had no effect (21). Neither gene had any effect on the rate of TA 3 GC transversions. Thus differences in effects on specific base pair mutations more than 1000-fold can be measured, which would be extremely difficult to do by sequencing. The necessity of using diploid strains for the assay is somewhat of a disadvantage, although a common strain, B-7462, is used in making all diploids, reducing the work involved in strain construction. Another potential disadvantage is that base pair mutation can be context dependent, and only one sequence context is being measured in this assay. However, Hampsey noted that other positions of the CYC1 gene are likely amenable to similar mutagenesis, which would give a different sequence context for the mutagenesis. We are in the process of making some of these additional mutants and also moving the CYC1 gene to a different chromosomal
MUTAGENESIS ASSAYS IN YEAST
position, to determine the extent of various contexts on mutation rates.
SUMMARY Once it is known that certain compounds or genetic conditions are mutagenic, the next step is to determine the specificity of the mutagenesis. This usually involves sequencing of independent isolates of a mutagenized target gene. Even with the ease of DNA sequencing, it is a substantial amount of work to generate a full mutagenic spectrum, particularly if there is a strong bias toward certain types of mutations. The assays presented here provide an attractive alternative. For conditions that cause frameshifts, the specificity for type of frameshift and the proportions of insertions and deletions can be easily determined. Similarly, a mutagenic spectrum for base pair substitutions can be derived and even large differences in effects on different bases can be determined.
REFERENCES 1. Grenson, M., Mousset, M., Wiame, J. M., and Bechet, J. (1966) Biochim. Biophys. Acta 127, 325–338. 2. Marsischky, G. T., Filosi, N., Kane, M. F., and Kolodner, R. (1996) Genes Dev. 10, 407– 420.
119
3. Kunz, B. A., Ramachandran, K., and Vonarx, E. J. (1998) Genetics 148, 1491–1505. 4. Pierce, M. K., Giroux, C. N., and Kunz, B. A. (1987) Mutat. Res. 182, 65–74. 5. Lee, G. S., Savage, E. A., Ritzel, R. G., and von Borstel, R. C. (1988) Mol. Gen. Genet. 214, 396 – 404. 6. von Borstel, R. C., Quah, S.-K., Steinberg, C. M., Flury, F., and Gottlieb, D. J. C. (1973) Genetics Suppl. 73, 141–151. 7. Steele, D. F., and Jinks-Robertson, S. (1992) Genetics 132, 9 –21. 8. Greene, C. N., and Jinks-Robertson, S. (1997) Mol. Cell. Biol. 17, 2844 –2850. 9. Henderson, S. T., and Petes, T. D. (1992) Mol. Cell. Biol. 12, 2749 –2757. 10. Strand, M., Prolla, T. A., Liskay, R. M., and Petes, T. D. (1993) Nature 365, 274 –276. 11. Sia, E. A., Kokoska, R. J., Dominska, M., Greenwell, P., and Petes, T. D. (1997) Mol. Cell. Biol. 17, 2851–2858. 12. Wierdl, M., Dominska, M., and Petes, T. D. (1997) Genetics 146, 769 –779. 13. Strand, M., Earley, M. C., Crouse, G. F., and Petes, T. D. (1995) Proc. Natl. Acad. Sci. USA 92, 10418 –10421. 14. Wierdl, M., Greene, C. N., Datta, A., Jinks-Robertson, S., and Petes, T. D. (1996) Genetics 143, 713–721. 15. Tran, H. T., Keen, J. D., Kricker, M., Resnick, M. A., and Gordenin, D. A. (1997) Mol. Cell. Biol. 17, 2859 –2865. 16. Hampsey, M. (1991) Genetics 128, 59 – 67. 17. Petes, T. D., Greenwell, P. W., and Dominska, M. (1997) Genetics 146, 491– 498. 18. Sherman, F. (1991) Methods Enzymol. 194, 3–21. 19. Lea, D. E., and Coulson, C. A. (1949) J. Genet. 49, 264 –284. 20. Luria, S. E., and Delbru¨ck, M. (1943) Genetics 28, 491–511. 21. Earley, M. C., and Crouse, G. F. (1998) Proc. Natl. Acad. Sci. USA 95, 15487–15491.
Mutagenesis Assays in Yeast Gray F. Crouse Department of Biology, Emory University, Atlanta, Georgia 30322
Analyzing mutation spectra is a very powerful method to determine the effects of various types of DNA damage and to understand the workings of various DNA repair pathways. However, compiling sequence-specific mutation spectra is laborious; even with modern sequencing technology, it is rare to obtain spectra with more than several hundred data points. Two assay systems are described for yeast, one for insertion/deletion mutations and one for base substitution mutations, that allow determination of specific mutations without the necessity of DNA sequencing. The assay for insertion/deletion mutations uses a variety of different simple repeats placed in frame with URA3 such that insertions or deletions lead to a selectable Ura ⫺ phenotype; essentially all such mutations are in the simple repeat sequence. The assay for base substitution mutations uses a series of six strains with different mutations in one essential codon of the CYC1 gene. Because only true reversions lead to a selectable phenotype, the bases mutated in any reversion event are known. The advantage of these assays is that they can quantitatively determine over several orders of magnitude the types of mutations that occur under a given set of conditions, without DNA sequencing. © 2000 Academic Press
Powerful genetics, ease of manipulation, and availability of the complete nucleotide sequence make Saccharomyces cerevisiae an excellent organism for the study of DNA damage and repair. One of the best methods for initial characterization of genes involved in DNA damage and repair is to analyze the spectrum of mutations created in their absence. Similarly, the mutagenic effects of external agents can be measured by analyzing their mutation spectra. A number of different systems in yeast have been used to assay mutations. One of the simplest and most commonly used assays is selection for inactivation of the CAN1 gene by growth on canavanine (1). The tacit assumption has been that a majority of inactivating mutations in the CAN1 gene would normally be due to base substitution errors, although recent sequencing of can1 mutations has revealed that in the absence of mismatch repair, a 116
majority of the mutations are frameshift mutations (2). The SUP4-o gene has been used as a mutational target in many studies from the Kunz laboratory (3, 4), and the URA3 gene has been used as a target in several laboratories (3, 5). Frameshift mutations have been studied by selection for reversion of the frameshift mutations hom3–10 (6) and lys2⌬Bgl (7). The difficulty with all of these assays is that to determine a mutational spectrum it is necessary to sequence each mutant gene. Even with modern sequencing methods, the number of events that can be analyzed is quite limited. Measurement of forward mutation rates in a gene generally requires that the entire gene be sequenced, which is a considerable task for the CAN1 gene, approximately 1.8 kb in size, but much easier for the small SUP4-o gene. The frameshift mutations mentioned above appear to revert mainly via compensating internal frameshift mutations which are limited to a region surrounding the original mutation that will restore the open reading frame and not interfere with function. Reversion of the lys2⌬Bgl mutation has been extensively studied and most reversion events occur within a 150-bp window (8); there is also a relatively short reversion window for the hom3–10 mutation, although the exact length over which compensating frameshift mutations can occur has not been as extensively explored (2). The article describes the use of two assays in yeast that can examine both frameshift (insertion/deletion) mutations and base pair substitutions with great specificity without sequencing. Most of the insertion/ deletion mutations observed in wild-type strains occur in repeated stretches of DNA and are apparently due to slipped mispairing (3, 8). Thus an assay designed to examine slipped mispairing mutations can serve as an assay for the general occurrence of frameshift or insertion/deletion mutations. A convenient assay for detecting simple repeat instability has been developed in the Petes laboratory (9 –12). This assay depends on a slippage event in a tester region either putting Escherichia coli lacZ in-frame and thus forming blue colo1046-2023/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
MUTAGENESIS ASSAYS IN YEAST
nies or shifting a URA3 gene out of frame, resulting in colonies resistant to 5-fluoroorotic acid (5FOA). The tester region can range from mononucleotide repeats to much longer repeats, but cannot be a multiple of 3. The assay with the URA3 gene is more convenient, since only cells with a slippage event will survive, and this assay has been used to examine the effects of mismatch repair, transcription, repeat length, and repeat size on insertion/deletion mutations (10 –14). A plating assay reveals the colonies resulting from a slippage event, but does not directly give information about the type of slippage event. However, a simple polymerase chain reaction (PCR) can determine the length of the tester region, and this information is sufficient to reveal the type of slippage event that occurred (10). A slippage assay has also been developed by placing various mononucleotide runs in the LYS2 gene, with most of the LYS2 mutations occurring in the added sequences (15). This assay system shares a number of characteristics with the one described here, but does not yet have a variety of substrates available. Determination of base pair substitution mutations has traditionally relied on sequencing of mutant genes obtained in a forward mutation assay, and unlike the frameshift reversion assays, it is necessary to sequence the entire gene for each individual mutation. However, Hampsey developed a tester system for yeast that can detect all base pair substitutions with a simple plating assay, and this assay system makes it possible to develop mutation spectra for base pair substitutions without sequencing (16). This assay uses six strains with a different base pair mutation in either position 1 or 2 of the essential codon Cys-22 of the CYC1 gene. Only a true reversion will enable yeast to use glycerol and thus a selective plating on glycerol plates will reveal the reversion rate for a specific base pair mutation.
MEASUREMENT OF INSERTION/DELETION MUTATION RATES All tester plasmids are derivatives of the pSH44 plasmid (9). An extensive series of test sequences differing in composition, length, and length of repeat in this vector have been made in the Petes laboratory (9, 11, 12, 17). The advantage of having the test sequence on a plasmid is that the plasmid can then be easily moved into different genetic backgrounds. A wide variety of genetic backgrounds can be used, but the system imposes certain constraints. The plasmid contains a TRP1 marker and selection is for a Ura ⫺ phenotype. In addition, the URA3 gene on the plasmid is under control of a LEU2 promoter, which requires growth in the absence of leucine and threonine for derepression of the promoter. Therefore the host strain must be trp1 ura3 Leu ⫹ Thr ⫹.
117
As with any mutation rate measurement, it is necessary to determine the number of mutants in many separate cultures. Instability of the test sequence in a plasmid is high enough in most cases that measurements can be made directly from colonies grown on plates without having to grow a liquid culture (12), thus decreasing the amount of effort involved. With stable test sequences, it is necessary to grow independent cultures in liquid to obtain enough cells for measurement. The procedure for mutation determination is as follows (12): Cells containing the test plasmid are plated on solid medium lacking only tryptophan. Preparation of medium is by standard procedures (18). After 3 days of growth at 30°C entire colonies are suspended in water and appropriate dilutions are plated on solid medium containing 0.1% 5FOA and lacking leucine, threonine, and tryptophan; viable cells are determined by plating dilutions on the same medium lacking 5FOA. Colonies are counted after incubation at 30°C for 3 days. Mutation rates are then determined using the method of the median (19). Because the test region is almost always mutated by insertion or deletion of various numbers of repeat units or by a mutation in the URA3 gene, the type of mutation can be revealed by determining the length of the altered region. Changes in length reflect a slippage event, and mutants without a change in length have usually undergone a mutation in the URA3 gene, which can be verified by sequencing (11). Length determination is most easily done by performing PCR on colonies or isolated genomic DNA using the primers 5⬘-CCAATAGGTGGTTAGCAATCG and 5⬘-GTTTTCCCAGTCACGAC and measuring the size of the resulting PCR product on a DNA sequencing gel (12). If necessary the PCR product can be sequenced directly. The advantage of this assay for determination of insertion/deletion mutation rates is that the specificity of insertion/deletion mutation can be determined without sequencing. For example, for a mononucleotide repeat, the ratio of rate of mutation in an msh2 strain to that in an msh3 strain is 48, whereas the same ratio for a tetranucleotide repeat is 1.2 (11). The ratio of deletions to insertions for the tetranucleotide repeat in an msh2 strain is 2.0, whereas the ratio for an msh3 strain is 3.5 (11). Thus from this assay one learns that MSH3 is relatively more active on a tetranucleotide loop than on a single-base-pair loop and that there is a difference in the action of MSH2 and MSH3 on a tetranucleotide loop. New target sequences can also be easily generated. For example, the mononucleotide repeat that has been used is a mononucleotide GC repeat (11), but the other three possible mononucleotide sequences could be used as well, which would test the effect not only on AT pairs, but on the strandedness of the sequences. One disadvantage of this form of the assay is that mutations in the URA3 gene will also lead to 5FOA
118
GRAY F. CROUSE
resistance. Resistance due to mutation is seen in as many as half of the 5FOA-resistant colonies from msh6 strains containing repeats inefficiently recognized by MSH6. Mutations in the URA3 gene are easily recognized as the most likely cause of 5FOA-resistant strains that do not show a change in repeat length. If necessary, such background can be eliminated by the use of an out-of-frame lacZ gene that is put in-frame by a change in repeat length (9). It should also be noted that although the assay is much easier to use when the test region is contained on a plasmid, the mutation rate is much lower when the sequence is in the chromosome and it is not clear that the mutational spectra are the same in the two states (10). Therefore in some cases, it may be desirable to put the test sequence in the chromosome (9, 13, 14).
MEASUREMENT OF BASE PAIR SUBSTITUTION RATES The function of the yeast CYC1 gene is absolutely dependent on having a cysteine residue at amino acid 22 of the protein. Strains mutant at Cys-22 will not grow on nonfermentable carbon sources and reversion of such strains can be selected by growth on glycerol. All revertants must restore the Cys at position 22; therefore for a given mutant strain, the mutation that restores function is known (16). A set of six strains that can together detect all possible base substitutions is available in an isogenic background (16). There are several important aspects of the genetic background of this tester system. The assay strains need to be diploid. It appears that if haploid strains are used, various recessive mutations can accumulate and lead to growth phenotypes that give irreproducible results (16). The diploid cells are made from a haploid cell containing the Cys-22 mutation and from a common haploid of opposite mating type. Both haploids have to be deleted for the related iso-2-cytochrome c gene (CYC7), or otherwise overexpression of CYC7 or recombination between CYC1 and CYC7 could lead to a Cyc ⫹ phenotype. In addition, the haploid cell of opposite mating type must be deleted for the CYC1 gene, resulting in diploids that are hemizygous for the Cys-22 mutant genes. All of the diploids are ura3, allowing the URA3 gene to be used as a selectable marker for gene disruptions, and the haploids of opposite mating type have either a leu2 or his1 marker, allowing easy selection for diploids. The assay procedure is as follows. The haplid strains containing the Cys-22 mutation are YMH2-7 (MAT␣ cyc1-22 ura3-52 leu2-3 112cyh2 cyc7-67), with the only difference between the six strains being the exact mutation in Cys-22, and B-7462 (MATa cyc1-1 cyc7-67 ura3-52 his1-1 can1-100). The cyc1-1 and cyc7-67 mutations are complete deletions. Any desired genetic
modifications are made on the haploid strains and then one of the YMH strains is mated with B-7462 by mixing the two strains, incubating on YPD plates (18) at 30°C overnight, and then selecting diploids by replica plating on SC minimal plates (18) supplemented with uracil. After purification of the diploids, all strains are grown for 2 days in YPD medium at 30°C to approximately 7 ⫻ 10 7 cells/ml. Five-milliliter cultures are harvested, washed with sterile doubly distilled H 2O, and resuspended to 1 ml in doubly distilled H 2O. A portion of each culture is diluted in doubly distilled H 2O and plated on two YPD plates to determine the number of cells in the culture. For most cultures, the remainder of the culture is plated onto YPGD plates, containing 3% (v/v) glycerol and 0.1% (w/v) glucose (16), at a density of no more than 9 ⫻ 10 3 cells/mm 2. For some cultures with an increased reversion rate, a tenfold dilution of the original culture is plated onto two YPGD plates at a density of no more than 7 ⫻ 10 2 cells/mm 2. YPD plates are counted after 2 days at 30°C and YPGD plates are counted after 7 days at 30°C. If cells are to be treated with mutagens, they are harvested and treated just before selective plating (16). Reversion rates are determined using a Luria and Delbru¨ck fluctuation analysis (20). The reversion rate of the unmodified diploid strains is very low, ranging between 10 ⫺10 and 10 ⫺9. To measure rates this low, it is necessary to plate all of the cells in each culture, and then use the number of cultures giving no colonies to determine the rate (20). A 5-ml culture can be plated on a total of three 150-mm Petri plates, without having problems with excessive density of cells. The advantage of this assay, in addition to eliminating the need for sequencing, is that very large differences in effects can be measured. For example, using the diploid strain YMH54, containing a GC substitution for a TA pair, it was found that deleting the mismatch repair gene MSH2 increased the rate of GC 3 TA transversions more than 2000 fold, whereas the MSH3 gene had no effect (21). Neither gene had any effect on the rate of TA 3 GC transversions. Thus differences in effects on specific base pair mutations more than 1000-fold can be measured, which would be extremely difficult to do by sequencing. The necessity of using diploid strains for the assay is somewhat of a disadvantage, although a common strain, B-7462, is used in making all diploids, reducing the work involved in strain construction. Another potential disadvantage is that base pair mutation can be context dependent, and only one sequence context is being measured in this assay. However, Hampsey noted that other positions of the CYC1 gene are likely amenable to similar mutagenesis, which would give a different sequence context for the mutagenesis. We are in the process of making some of these additional mutants and also moving the CYC1 gene to a different chromosomal
MUTAGENESIS ASSAYS IN YEAST
position, to determine the extent of various contexts on mutation rates.
SUMMARY Once it is known that certain compounds or genetic conditions are mutagenic, the next step is to determine the specificity of the mutagenesis. This usually involves sequencing of independent isolates of a mutagenized target gene. Even with the ease of DNA sequencing, it is a substantial amount of work to generate a full mutagenic spectrum, particularly if there is a strong bias toward certain types of mutations. The assays presented here provide an attractive alternative. For conditions that cause frameshifts, the specificity for type of frameshift and the proportions of insertions and deletions can be easily determined. Similarly, a mutagenic spectrum for base pair substitutions can be derived and even large differences in effects on different bases can be determined.
REFERENCES 1. Grenson, M., Mousset, M., Wiame, J. M., and Bechet, J. (1966) Biochim. Biophys. Acta 127, 325–338. 2. Marsischky, G. T., Filosi, N., Kane, M. F., and Kolodner, R. (1996) Genes Dev. 10, 407– 420.
119
3. Kunz, B. A., Ramachandran, K., and Vonarx, E. J. (1998) Genetics 148, 1491–1505. 4. Pierce, M. K., Giroux, C. N., and Kunz, B. A. (1987) Mutat. Res. 182, 65–74. 5. Lee, G. S., Savage, E. A., Ritzel, R. G., and von Borstel, R. C. (1988) Mol. Gen. Genet. 214, 396 – 404. 6. von Borstel, R. C., Quah, S.-K., Steinberg, C. M., Flury, F., and Gottlieb, D. J. C. (1973) Genetics Suppl. 73, 141–151. 7. Steele, D. F., and Jinks-Robertson, S. (1992) Genetics 132, 9 –21. 8. Greene, C. N., and Jinks-Robertson, S. (1997) Mol. Cell. Biol. 17, 2844 –2850. 9. Henderson, S. T., and Petes, T. D. (1992) Mol. Cell. Biol. 12, 2749 –2757. 10. Strand, M., Prolla, T. A., Liskay, R. M., and Petes, T. D. (1993) Nature 365, 274 –276. 11. Sia, E. A., Kokoska, R. J., Dominska, M., Greenwell, P., and Petes, T. D. (1997) Mol. Cell. Biol. 17, 2851–2858. 12. Wierdl, M., Dominska, M., and Petes, T. D. (1997) Genetics 146, 769 –779. 13. Strand, M., Earley, M. C., Crouse, G. F., and Petes, T. D. (1995) Proc. Natl. Acad. Sci. USA 92, 10418 –10421. 14. Wierdl, M., Greene, C. N., Datta, A., Jinks-Robertson, S., and Petes, T. D. (1996) Genetics 143, 713–721. 15. Tran, H. T., Keen, J. D., Kricker, M., Resnick, M. A., and Gordenin, D. A. (1997) Mol. Cell. Biol. 17, 2859 –2865. 16. Hampsey, M. (1991) Genetics 128, 59 – 67. 17. Petes, T. D., Greenwell, P. W., and Dominska, M. (1997) Genetics 146, 491– 498. 18. Sherman, F. (1991) Methods Enzymol. 194, 3–21. 19. Lea, D. E., and Coulson, C. A. (1949) J. Genet. 49, 264 –284. 20. Luria, S. E., and Delbru¨ck, M. (1943) Genetics 28, 491–511. 21. Earley, M. C., and Crouse, G. F. (1998) Proc. Natl. Acad. Sci. USA 95, 15487–15491.