Mutations induced at the white and vermilion loci in Drosophila melanogaster

The white and vermilion loci in D. melanogaster were selected as target genes for the study of the mutational specificity of ionizing radiation and N-ethyl-N-nitrosourea (ENU) in a whole organism. Analysis of X-ray- and neutron-induced white mutants by a combination of genetic and molecular techniques showed that ionizing radiation induces primarily break-type mutations against a repair-proficient background, the majority of these alterations being deletions. Both very large multi-locus deficiencies and deletions of only a few base pairs were observed. These small deletions are flanked by repeats of 2-3 nucleotides, one copy of which is retained at the new junction. Presumably these small repeats are involved in the generation of the X-ray-induced deletions. In excision-repair-deficient mus201D1 flies, the frequency of whole-body white mutants recovered after X-ray irradiation is the same as in the wild-type strain. The percentage of mosaic mutations, however, is enhanced by a factor 3-4. Analysis by blot hybridization of ENU-induced white mutants strongly indicates that most mutations are due to base-pair changes. This was confirmed by sequence analysis of 25 ENU-induced vermilion mutants. In all mutants the alterations are due to base-pair changes, the majority being GC to AT transitions (61%).     

X-ray-induced specific-locus mutations in the ad-3 region of two-component heterokaryons of Neurospora crassa. IV. Irreparable mutants of genotype ad-3A and ad-3B result from multilocus deletion and an unexpectedly high frequency of multiple-locus mutations

The induction of specific-locus mutations in the ad-3 region of Neurospora crassa after X-irradiation was studied in a two-component heterokaryon to determine: (1) the ratio of reparable ad-3 mutants (presumed gene/point mutations, designated ad-3R) to irreparable ad-3 mutants (presumed multilocus deletions, designated ad-3IR), and (2) the induction kinetics of each class (Webber and de Serres, 1965). More extensive genetic tests made subsequently (de Serres, 1989a) on the 832 X-ray-induced specific-locus mutations recovered in those experiments showed that unexpected high frequencies of reparable and irreparable ad-3 mutants are actually multiple-locus mutants that have additional, but separate, sites of recessive lethal (RLCL) damage in the immediately adjacent genetic regions (designated ad-3R + RLCL or ad-3IR + RLCL). The frequencies of these X-ray-induced multiple-locus mutants in the ad-3 region are orders of magnitude higher than expected on the basis of target theory (where the frequency of the double mutant is expected to be the product of the frequencies of each single mutant) and classical models of chromosome structure during interphase (de Serres, 1989a). In the present paper, a random sample of 832 X-ray-induced ad-3 mutants of genotype ad-3A or ad-3B that are irreparable have been subjected to more extensive genetic fine-structure analysis. These experiments were designed to determine the extent of the functional inactivation in individual mutants in the ad-3 and immediately adjacent genetic regions in mutants classified as presumptive multilocus deletions or multiple-locus mutations. These experiments have shown that in Neurospora crassa most X-ray-induced irreparable mutants of genotype ad-3A or ad-3B map as a series of overlapping multilocus deletions. Among the 29 irreparable mutants of genotype ad-3A, there are 16 different subgroups of complementation patterns; and among the 63 irreparable mutants of genotype ad-3B, there are also 16 different subgroups. In addition, mutants classified as presumptive multiple-locus mutants result from a variety of separate, but closely linked, sites of genetic damage.(ABSTRACT TRUNCATED AT 400 WORDS)     

X-ray-induced specific-locus mutations in the ad-3 region of two-component heterokaryons of Neurospora crassa. VIII. Dose-dependence of the overall spectrum

There is considerable controversy in the literature concerning the nature of X-ray-induced specific-locus mutations in various experimental organisms. To investigate this problem in Neurospora crassa a series of experiments (Webber and de Serres, 1965) was performed to study the induction-kinetics of X-ray-induced mutation in the adenine-3 (ad-3) region of a two-component heterokaryon (H-12). Subsequent genetic analyses (de Serres, 1989a,b,c, 1990a), on a series of 832 mutants recovered in these experiments, have shown that 3 different classes of ad-3 mutants were recovered, namely gene/point mutations, multilocus deletions and multiple-site mutations. Complementation studies with a series of genetic markers that define 21 genetic loci in the ad-3 and immediately adjacent genetic regions have shown that ad-3 mutants classified as multilocus deletions result from the inactivation of a series of loci in the ad-3 and immediately adjacent regions of Linkage Group I, whereas multiple-locus mutations result from combinations of gene/point mutations and multilocus deletions. Analysis of the induction kinetics of these 3 different classes, after completion of the genetic characterization of all mutants (de Serres, 1990b) demonstrated that gene/point mutations increase linearly with X-ray dose, whereas multilocus deletions and multiple-site mutations increase as the square of X-ray dose. Further analysis of allelic complementation among the gene/point mutations at the ad-3B locus (de Serres, 1990c), demonstrated that the spectrum of complementation patterns was dose-dependent: complementing mutants with nonpolarized patterns decreased and noncomplementing mutations increased with increasing X-ray dose. There was little or no change with dose in the frequency of mutants with polarized patterns. In the present report, data from studies published previously have been utilized, along with additional data from the original X-ray experiments (12-5, 12-6, 12-7, and 12-10; see Webber and de Serres, 1965) to develop composite complementation maps of the X-ray-induced specific-locus mutations in the ad-3 and immediately adjacent regions as a function of X-ray dose. This analysis of the overall spectrum of X-ray-induced specific-locus mutations in the ad-3 region demonstrated marked dose-dependence and provides an explanation for the discrepancies in the literature with regard to specific-locus studies in different experimental organisms.     

Molecular analysis of X-ray-induced mutants at the HPRT locus in V79 Chinese hamster cells

Spontaneous and X-ray-induced mutants at the hypoxanthine phosphoribosyl transferase (HPRT) locus have been isolated from V79 Chinese hamster cells and characterized at the biochemical and cytogenetic levels. Fourteen spontaneous and 24 X-ray-induced clones were azaguanine and thioguanine resistant, did not grow in HAT medium (AZRTGRHATS) and failed to incorporate significant levels of [14C]hypoxyanthine. Cytogenetic analysis of two spontaneous and eight X-ray-induced mutants revealed no major X chromosome rearrangements. In two induced mutants, one of which was hypotetraploid (mode 35-39) with 2 X chromosomes, the short arm of the chromosome (Xp) was slightly shorter than normal. A third mutant was hyperdiploid (mode 22-23) compared with the parental clone (mode 21). When compared with wild-type clones, no other cytogenetic changes were evident in the remaining mutants. Analysis at the DNA level using a Chinese hamster HPRT cDNA probe showed major deletion of HPRT sequences in two and partial deletion in another two induced mutants. In two of the mutants with deletions of HPRT sequences there was a visible shortening of the Xp arm. In the other six mutants two spontaneous and four induced) no karyotypic changes or alterations in restriction fragment patterns were detected suggesting that they carry small deletions or point mutations at the HPRT locus.     

Rescue of adeno-associated virus from recombinant plasmids: gene correction within the terminal repeats of AAV

We have isolated three types of pBR322-AAV recombinant plasmids that contain deletions within the 145 bp AAV terminal repeats. When the plasmids were transfected into human cells, mutants that contained deletions within the left (type I) or right (type II) terminal repeat were viable. Of four mutants examined that contained deletions in both termini (type III), only one was viable. All of the viable mutants produced AAV virions that contained wild-type AAV DNA. Furthermore, the viable type III deletion could be converted to a nonviable mutant by deleting all copies of an 11 bp sequence from its termini. We conclude that there is an efficient mechanism for correcting deletions within the AAV termini. A model that could account for these observations is also discussed.     

DNA sequence alterations in Hr-t deletion mutants of polyoma virus.

We have investigated the DNA sequence alterations in several hr-t mutants of polyoma virus. These mutants are defective in one of the two known viral functions essential for transformation and are altered with respect to several minor T antigen species. The lesions in some of these mutants have been mapped previously by marker rescue experiments to Hpa II fragment 4 (Hpa II-4, 78.4--91.7 map units) in the proximal part of the early region of the viral DNA. Thirteen of sixteen hr-t mutants examined carry deletions 2 to 5 map units (100--250 bp) long in Hpa 11-4. Three mutants carry either point mutations or very small deletions/insertions. Eight of the deletion mutants were mapped closely with restriction enzymes. Seven of them have deletions located entirely within the Hae III subfragment A of Hpa II-4 (the Hae A subfragment, 78.4--85.2 map units), and one extends just beyond this subfragment, ending at 85.5 map units. The complete sequence of the wild-type Hae A subfragment was determined and compared with those of four deletion mutants, NG-18, A-8, 6B5 and B-2. The deletion in each of these mutants is out-of-phase: NG-18, 187 bp; A-8, 127 bp; 6B-5, 179 bp; B-2, 241 bp. All are expected to remove protein sequences in the C terminal part of the small t antigen.     

X-ray-induced specific-locus mutations in the ad-3 region of two-component heterokaryons of Neurospora crassa. II. More extensive genetic tests reveal an unexpectedly high frequency of multiple-locus mutations

More extensive genetic tests have been performed on a series of 832 X-ray-induced specific-locus mutations in the ad-3 region of a 2-component heterokaryon (H-12) of Neurospora crassa, reported earlier (Webber and de Serres 1965). Using a new tester strains and techniques for performing large-scale genetic tests (heterokaryon, dikaryon and trikaryon) to characterize ad-3 mutants induced in 2-component heterokaryons, new data have been obtained on this sample of X-ray-induced ad-3 mutants. These new data show that unexpectedly high frequencies of both single-locus (gene/point) mutations and multilocus deletions in the ad-3 region have additional, but separate, sites of resessive lethal (RL^CL) damage in the immediately adjacent genetic regions. The frequencies of these X-ray-induced multiple-locus mutants in the ad-3 region are orders of magnitude higher than expected on the basis of target theory and classical models of chromosome structure during interphase. Current models of interphase chromosome structure in higher eukaryotes as revealed by chromosome “painting” offer a possible explanation of the Neurospora data.     

Mutations induced by X-rays at the HPRT locus in cultured Chinese hamster cells are mostly large deletions

We investigated the molecular basis of 19 X-ray-induced HPRT-deficient mutants of V79 Chinese hamster cells with Southern hybridisation techniques. 12 of those mutants suffer from a big deletion (greater than 10 kb) of HPRT DNA sequences. Cytological studies of chromosome preparations of those 12 deletion mutants showed that in at least 3 of these mutants part of the long arm of the X-chromosome was lost. After correction for spontaneous arising mutations we estimate that at least 70-80% of X-ray-induced mutations are caused by large deletions.     

Deletion mutagenesis independent of recombination in bacteriophage T7.

Deletion between directly repeated DNA sequences in bacteriophage T7-infected Escherichia coli was examined. The phage ligase gene was interrupted by insertion of synthetic DNA designed so that the inserts were bracketed by 10-bp direct repeats. Deletion between the direct repeats eliminated the insert and restored the ability of the phage to make its own ligase. The deletion frequency of inserts of 85 bp or less was of the order of 10(-6) deletions per replication. The deletion frequency dropped sharply in the range between 85 and 94 bp and then decreased at a much lower rate over the range from 94 to 900 bp. To see whether a deletion was predominantly caused by intermolecular recombination between the leftmost direct repeat on one chromosome and the rightmost direct repeat on a distinct chromosome, genetic markers were introduced to the left and right of the insert in the ligase gene. Short deletions of 29 bp and longer deletions of approximately 350 bp were examined in this way. Phage which underwent deletion between the direct repeats had the same frequency of recombination between the left and right flanking markers as was found in controls in which no deletion events took place. These data argue against intermolecular recombination between direct repeats as a major factor in deletion in T7-infected E. coli.     

X-ray-induced specific-locus mutations in the ad-3 region of two-component heterokaryons of Neurospora crassa. VI. Induction kinetics of gene/point mutations, multilocus deletions and multiple-locus mutations

Genetic fine-structure analysis of X-ray-induced specific-locus mutants in the ad-3 region of two-component heterokaryons of Neurospora crassa has shown that gene/point mutations, multilocus deletions and multiple-locus mutations are induced. When the dose-response curves for these classes of ad-3 mutants were plotted, it was demonstrated that X-ray-induced gene/point mutations (ad-3R) increased linearly with X-ray dose and X-ray-induced multilocus deletions increased as the square of the X-ray dose. However, all classes of multiple-locus mutations, which would be expected to result from 3 to 8 hits on the basis of target theory (Lea, 1955), were found to increase as the square of the dose. Target theory assumes that the DNA of individual chromosomes is distributed randomly throughout the interphase nucleus. A model of eukaryotic interphase chromosome structure in which the DNA of individual chromosomes presents a nonrandom target to X-rays [Pinkel et al., Proc. Natl. Acad. Sci. (U.S.A.) 83 (1986), 2934-2938] provides a possible explanation for the high frequency and dose-squared induction kinetics of the multiple-locus mutants induced by X-rays in the ad-3 region.     

X-ray-induced specific-locus mutations in the ad-3 region of two-component heterokaryons of Neurospora crassa. V. Irreparable mutants of genotype ad-3A ad-3B, ad-3A ad-3B nic-2, and ad-3B nic-2 result from multilocus deletion and an unexpectedly high frequency of multiple-locus mutations

More extensive complementation tests than those performed initially (Webber and de Serres, 1965) on a series of 832 X-ray-induced specific-locus mutations in the adenine-3 (ad-3) region of a two-component heterokaryon (H-12) of Neurospora crassa (de Serres, 1989a) showed that unexpectedly high frequencies of specific-locus mutations in the ad-3 region have additional, but separate, sites of recessive lethal (RLCL) damage in the immediately adjacent genetic regions. The frequencies of these X-ray-induced multiple-locus mutants in the ad-3 region are orders of magnitude higher than that expected on the basis of target theory and classical models of chromosome structure during interphase (de Serres, 1989a). Genetic fine structure analyses, by means of homology tests with tester strains carrying genetic markers in the ad-3 and immediately adjacent regions, have been performed to map the presumed multiple-locus mutations. In a previous paper (de Serres, 1989c), X-ray-induced irreparable ad-3 mutants of the following genotypes and numbers (ad-3A or ad-3B were analyzed, and the high frequency of multiple-locus mutations was confirmed. In the present paper, X-ray-induced irreparable ad-3 mutants of the following genotypes and numbers (ad-3A ad-3B, ad-3A ad-3B nic-2, and ad-3B nic-2 have also been subjected to the same genetic fine structure analysis. These experiments, in the previous (de Serres, 1989c) and present papers, were designed to determine the extent of the functional inactivation in the ad-3 and immediately adjacent genetic regions in individual mutants classified as presumptive multilocus deletions or multiple-locus mutations.     

Radiation specific patterns of loss of heterozygosity on chromosome 17q

We and others have previously reported that the percentage of ionizing radiation-induced TK(-) mutants exhibiting loss of heterozygosity (LOH) is not significantly different from those occurring spontaneously. In order to search further for a distinguishing feature of the X-ray-induced spectrum, and to characterize mechanisms of chromosomal scale mutagenesis, we used detailed mapping information to analyze the extent of LOH along chromosome 17q. Significant differences were observed when the extent of LOH tracts was considered. The representation of very long LOH tracts (>/=41 cM) was significantly (p=0.004) more common among spontaneous mutants, while relatively local LOH events, involving only markers in a 1-10 cM region surrounding the tk locus, are significantly (p=0.018) more prevalent among X-ray-induced mutants. Our data suggests that, although large deletions are recoverable, X-ray-induced autosomal deletions are not evenly distributed over the available size range. This indicates a mechanistic rather than biological restriction to the size of radiation-induced deletions, and demonstrates that the pattern of LOH may also be useful as a distinguishing component of the mutational spectrum.     

Molecular and biochemical analyses of spontaneous and X-ray-induced mutants in human lymphoblastoid cells

We have isolated a series of 14 spontaneously arising and 28 X-ray-induced mutants at the hypoxanthine-guanine phosphoribosyltransferase (hgprt) locus in human lymphoblastoid cells. Among the spontaneous mutants, 5/14 (36%) had detectable alterations in their restriction fragment pattern after hybridization with a human cDNA probe for hgprt. Of the 10 remaining mutants, 4 had partial HGPRT enzyme activity, which suggested that they contained point mutations. Among the 28 mutants induced by 150 rad of X-rays, 15 (54%) had deletions of part or all of the hgprt gene. 5 of the remaining 13 (18% overall) had partial HGPRT enzyme activity, which suggested that they contained point mutations. These data imply that in this human cell system, X-rays induce both point mutants which have residual enzyme activity as well as mutations involving relatively large deletions of DNA.     

The nature of radiation-induced mutations at the white locus of Drosophila melanogaster

X-Ray- and neutron-induced mutations at the white locus of Drosophila melanogaster were used to study the nature of radiation-induced genetic damage. Genetic analysis showed the presence of multi-locus deficiencies in 15 out of 31 X-ray mutants and in 26 out of 35 mutants induced by neutrons. The DNA from 11 X-ray and 4 neutron mutants, which were not multi-locus deficiencies, was analyzed by Southern blot-hybridization. Deletions were observed in 2 X-ray and 1 neutron mutant. In combination with cytogenetic techniques, chromosomal rearrangements affecting the white locus (translocations, inversions, etc.) were identified in 3 X-ray and in 2 neutron mutants. A hot-spot for translocation breakpoints was identified in the left arm of the third chromosome. 5 X-ray mutants, which apparently did not contain large deletions, were subjected to further analysis by the nuclease S1 protection method, after cloning of the white gene. In 4 mutants a small deletion could indeed be detected in this way. Thus it seems that by far the main part of X-ray- and neutron-induced white mutants have arisen through large changes in the white gene, especially deletions.     

Two distinct mechanisms for deletion in mitochondrial DNA of Schizosaccharomyces pombe mutator strains. Slipped mispairing mediated by direct repeats and erroneous intron splicing

Mutator strains of the fission yeast Schizosaccharomyces pombe produce mitochondrial respiratory deficient mutants at a high rate, and roughly 20% of these mutants carry deletions in the range of 50 to 1500 base-pairs. To elucidate the mechanism of deletion we have sequenced ten deletion mutants in the mosaic gene encoding apocytochrome b (cob) and three in the split gene coding for the first subunit of cytochrome c oxidase (cox1). Of 13 deletions, ten are correlated with the presence of direct repeats, which could promote deletions by slipped mispairing during DNA replication. In some of these mutants, the termini are located in possible DNA secondary structures. In three independently isolated mutants with identical deletions in the cob gene, the 5' deletion endpoint coincides with the 3' splice point of the intron, whereas the 3' endpoint of the deletion exhibits pronounced homology with the 5' splice point of the intron. This result suggests that these deletions might be initiated by erroneous RNA splicing.     

X-ray-induced specific-locus mutations in the ad-3 region of two-component heterokaryons of Neurospora crassa. XI. Heterozygous effects of gene/point mutations of genotype ad-3A or ad-3B.

Previous studies on X-ray-induced irreparable adenine-3 mutants (designated ad-3IR), induced in heterokaryon 12 of Neurospora crassa, showed that they were not recessive, and that they demonstrated heterozygous effects in terms of markedly reduced linear growth rates as compared with a wild-type control (de Serres, 1965, 1988). Homology tests on X-ray-induced irreparable mutants showed that they map, in the main part, as a series of overlapping multilocus deletions that extend both proximally and distally into the immediately adjacent genetic regions, as well as into the 'X' region (a region of unknown, but essential, function) between ad-3A and ad-3B (de Serres, 1969, 1989a). Studies on a larger sample of X-ray-induced multilocus deletion mutations of genotype (ad-3A)IR or (ad-3B)IR (de Serres et al., 1992) demonstrated that heterozygous effects are allele specific and that there was no correlation with genotype, radiation dose or complementation map position. Furthermore, the heterozygous effects of multilocus deletions in the ad-3 region can be modified genetically and biochemically (de Serres and Miller, 1988). In the present paper, the heterozygous effects of X-ray-induced gene/point mutations of genotype ad-3AR or ad-3BR, induced in heterokaryon 12 (Webber and de Serres, 1965; de Serres, 1988, 1989a), were determined. The studies presented in this paper show that 8.1% (3/37) of X-ray-induced ad-3AR mutations exhibit heterozygous effects in terms of reduced linear growth rates in forced dikaryons with a gene/point mutant at the ad-3B locus, and 10.8% (4/37) in forced dikaryons with a multilocus deletion mutation covering the ad-3B locus. In addition, 24.3% (9/37) of ad-3AR mutations exhibit heterozygous effects in terms of enhanced linear growth rates in forced dikaryons with a gene/point mutant at the ad-3B locus. Similar studies with X-ray-induced ad-3BR mutations showed that 54.9% (28/51) exhibit heterozygous effects in terms of reduced growth rates in forced dikaryons with a gene/point mutant at the ad-3A locus and 100.0% (48/48) in forced dikaryons with a multilocus deletion covering the ad-3A locus. These studies have also shown that about a 13-fold higher percentage of X-ray-induced multiple-locus mutations of genotype ad-3AR + RLCL have heterozygous effects resulting in reduced growth rates than X-ray-induced single-locus mutations of genotype ad-3AR. The overall data base on X-ray-induced ad-3 gene/point mutations in the present studies demonstrates that heterozygous effects are allele specific, genotype specific, and locus specific.(ABSTRACT TRUNCATED AT 400 WORDS)     

X-ray-induced specific-locus mutations in the ad-3 region of two-component heterokaryons of Neurospora crassa, IX. Mutational spectra as a function of X-ray dose.

In previous studies, X-ray-induced specific-locus mutations in the adenine-3 (ad-3) region of a two-component heterokaryon (H-12) of Neurospora crassa were combined with a series of tester strains carrying markers in the ad-3 and immediately adjacent regions to map mutants that were presumed multilocus deletions (de Serres, 1989c, 1990a). Two new classes of X-ray-induced mutations were recovered: multiple-locus mutations consisting of gene/point mutations at the ad-3A or ad-3B locus with a closely linked recessive lethal mutation, or multilocus deletions covering the ad-3A, ad-3B and/or nic-2 loci with a closely linked recessive lethal mutation (designated ad-3R + RLCL and [ad-3]IR + RLCL, respectively). Thus, the ad-3 specific-locus assay can detect damage occurring at the ad-3A and the ad-3B loci, as well as at a minimum of 19 other loci in the immediately adjacent regions. The original overall spectrum of ad-3 mutations can be resolved, by genetic analysis, into a series of 30 subclasses. In the present paper, the data from the genetic analysis of 832 X-ray-induced mutants recovered from a series of 4 experiments (Webber and de Serres, 1965) have been presented in terms of Mutational Spectra organized as a function of X-ray dose. Comparison of these Spectra demonstrates the shift from high percentages of gene/point mutations (with a high percentage of mutants at the ad-3B locus showing allelic complementation) at low doses, to low percentages of gene/point mutations (with a low percentage of ad-3B mutants showing allelic complementation) and high percentages of multilocus deletion mutations and multiple-locus mutations (of genotype ad-3R + RLCL or [ad-3]IR + RLCL) at high doses. These Mutational Spectra demonstrate the marked dose-dependence of X-ray-induced specific-locus mutations in a eukaryotic organism.     

X-ray-induced specific-locus mutations in the ad-3 region of two-component heterokaryons of Neurospora crassa. VII. Genetic lesions resulting in gene/point mutations at the ad-3B locus have different dose-response relationships

Genetic characterization of X-ray-induced ad-3 mutants, induced in a two-component heterokaryon (H-12) of Neurospora crassa, has been performed to determine genotype, patterns of allelic complementation, and leakiness, and to distinguish gene/point mutations from multilocus deletions and multiple locus mutations (de Serres, 1989c, 1990a). The array of genotypes in the classes and subclasses in the spectrum of ad-3 mutants induced by a mutagenic agent constitute its mutagenicity profile; for X-rays this profile consists of 29 different genotypes. In the present paper, the data on gene/point mutations induced at the ad-3B locus (designated ad-3BR mutations) have been tabulated as a function of X-ray dose to determine the dose-dependent relationships of complementing and noncomplementing mutants. This analysis has shown that the overall percentages of mutants showing allelic complementation and the percentages of complementing mutants with nonpolarized patterns (both leaky and nonleaky) and noncomplementing mutants were dose-dependent; the former class decreased with increasing X-ray dose, and the latter class increased with increasing X-ray dose. The percentages of complementing mutants with polarized patterns were X-ray dose-independent. In addition, an unexpectedly high frequency of mutants with nonpolarized complementation patterns are discontinuous and probably result from multiple-site mutations.     

Analysis of repeat-mediated deletions in the mitochondrial genome of Saccharomyces cerevisiae

Mitochondrial DNA deletions and point mutations accumulate in an age-dependent manner in mammals. The mitochondrial genome in aging humans often displays a 4977-bp deletion flanked by short direct repeats. Additionally, direct repeats flank two-thirds of the reported mitochondrial DNA deletions. The mechanism by which these deletions arise is unknown, but direct-repeat-mediated deletions involving polymerase slippage, homologous recombination, and nonhomologous end joining have been proposed. We have developed a genetic reporter to measure the rate at which direct-repeat-mediated deletions arise in the mitochondrial genome of Saccharomyces cerevisiae. Here we analyze the effect of repeat size and heterology between repeats on the rate of deletions. We find that the dependence on homology for repeat-mediated deletions is linear down to 33 bp. Heterology between repeats does not affect the deletion rate substantially. Analysis of recombination products suggests that the deletions are produced by at least two different pathways, one that generates only deletions and one that appears to generate both deletions and reciprocal products of recombination. We discuss how this reporter may be used to identify the proteins in yeast that have an impact on the generation of direct-repeat-mediated deletions.