High-resolution mapping of spontaneous mitotic recombination hotspots on the 1.1 Mb arm of yeast chromosome IV

Abstract

Although homologous recombination is an important pathway for the repair of double-stranded DNA breaks in mitotically dividing eukaryotic cells, these events can also have negative consequences, such as loss of heterozygosity (LOH) of deleterious mutations. We mapped about 140 spontaneous reciprocal crossovers on the right arm of the yeast chromosome IV using single-nucleotide-polymorphism (SNP) microarrays. Our mapping and subsequent experiments demonstrate that inverted repeats of Ty retrotransposable elements are mitotic recombination hotspots. We found that the mitotic recombination maps on the two homologs were substantially different and were unrelated to meiotic recombination maps. Additionally, about 70% of the DNA lesions that result in LOH are likely generated during G1 of the cell cycle and repaired during S or G2. We also show that different genetic elements are associated with reciprocal crossover conversion tracts depending on the cell cycle timing of the initiating DSB.

Figure 1. Patterns of gene conversion resulting from the repair of G1- or G2-generated DSBs.

Chromatids that are depicted in black represent YJM789-derived chromatids, and those in red represent W303a-derived chromatids. The centromeres are shown as circles. A. 3∶1 conversion event. Repair of a G2-associated DSB results in a 3∶1 gene conversion tracts (enclosed in dotted lines) associated with the LOH event. B. 4∶0 conversion event. A chromosome in a G1 cell is replicated to form two sister chromatids that are broken at approximately the same position. Repair of these two broken chromatids in G2 can result in a region in which all four chromatids have identical SNPs, a 4∶0 conversion tract. In this example, both conversion tracts are of the same length. C. 4∶0/3∶1 hybrid tract. As in Figure 1B, a broken chromosome is replicated to form to two broken sister chromatids. The conversion tracts associated with the repair of the two DSBs, however, are of different lengths, resulting in a 4∶0/3∶1 hybrid tract.

Figure 2. Mapping of a crossover with an associated 3∶1 conversion event by SNP microarrays.

The values on the Y-axis show the experimental/reference hybridization ratio of genomic DNA to oligonucleotides that are specific to SNPs from the W303a and YJM789 backgrounds. The values on the X-axis indicate the SGD coordinates of the SNPs along chromosome IV. The red and blue lines or data points depict the hybridization levels of probes specific for the W303a and YJM789 homologs, respectively. Where both the red and blue lines have a value near 1, the diploid is heterozygous for the SNPs. If the strain is homozygous for the W303a variant, the red signal increases to about 1.6 and the blue signal diminishes to about 0.3. If the strain is homozygous for the YJM789 variant, the reverse pattern is observed. A. Low-resolution depiction of a reciprocal crossover analyzed by the SNP microarrays. The top and bottom plots represent the red and white sectors, respectively. In both the red and white sectors, there is a transition between heterozygosity and homozygosity at approximately SGD coordinate 770 kb. B. High-resolution depiction of a reciprocal crossover. Each data point indicates the hybridization level to a specific oligonucleotide. Each square and diamond indicates the hybridization level to a specific oligonucleotide. The genomic DNA from the red sector has a transition between heterozygosity and homozygosity at approximately coordinate 762.5 kb (top panel), whereas genomic DNA from the white sector has a transition at about coordinate 772 kb. The regions shown in green rectangles represent the 3∶1 gene conversion event.

Figure 3. Crossovers and associated gene conversion events in JSC25 along the right arm of chromosome IV.

The X-axis represents the SGD coordinates (the centromere is at 449 kb), and the Y-axis represents the length of the individual conversion tract. The red and black horizontal lines in the center of the figure depict the W303a and YJM789 homologs, respectively. Individual conversion tracts are depicted as two parallel vertical lines that are attached to either homolog line by a black connector line which indicates the location of the centromere-proximal margin of the conversion tract. Conversion tracts depicted in the top portion of the figure and connected to the W303a homolog, resulted from DSBs initiated on the W303a homolog. Conversion tracts depicted in the lower half of the figure were initiated by DSBs on the YJM789 homolog. The vertical left and right members of the line pairs show the conversion tract of the red and white sides of the sector, respectively. Green, red, and black show heterozygosity, homozygosity for W303a SNPs, and homozygosity for YJM789 SNPs, respectively. Each segment of the depicted conversion tract is proportional to the size of that segment within the sector. The hash marks between the two homologs indicate the density of the SNPs represented on the microarray.

Figure 4. Summary of conversion tracts associated with crossovers along the right arm of chromosome IV.

These plots represent the number of times a SNP represented on the microarray is involved in a conversion tract. A SNP was considered involved if it was between the start and end of the conversion tract. A. All conversion tracts. The labels HS1–HS7 represent potential hotspots for recombination. B. Conversion tract distribution of the events initiated on the W303a (red) and YJM789 (black) homolog. C. Conversion tract distribution of events initiated during G1 (purple) or G2 (green line) of the cell cycle.

Figure 5. Comparison of physical and genetic mitotic recombination maps.

This figure shows a comparison of the physical map of the 1.1 Mb right arm of chromosome IV with a genetic map of mitotic crossovers. We show three genetic maps: all crossovers (JSC25 map), crossovers initiated from the W303a homolog (W303a map), and crossovers initiated from the YJM789 homolog (YJM789 map). The chromosome arm was divided into 20 regions of about 53 kb. The genetic distances are based on the proportion of crossovers within each 53 kb interval, using the “point-analysis” method employed for Figure S2 and described in Text S1. The top and bottom horizontal lines show the physical map. The vertical lines allow comparisons of the genetic maps with the physical maps, and of the different genetic maps with each other. The relative “thinness” and “thickness” of the same segment between the W303a and YJM789 maps indicates whether that region has more crossovers initiated on one homolog in comparison to the other. The recombination hotspots (labeled in Figure 4A) are indicated by rectangles on the physical map lines (top and bottom lines). The red, blue, and black colors indicate hotspots that are specific to the W303a-derived homolog, specific to the YJM789-derived homology, and hotspots that are found on both homologs, respectively.

Figure 6. Deletion analysis of the HS4 hotspot.

A. Experimental system used to monitor the activity of the wild-type HS4 hotspot and various deletion derivatives. The URA3 and HYG genes were inserted centromere-distal and -proximal to the inverted Ty elements, respectively, on the W303a-derived homolog (black). The YJM789-derived homolog (grey) was unaltered. B. Modifications of the inverted repeat structure. In the control strain JSC71-1, the unmodified hotspot consists of a Ty2 element on the Watson strand, 25 bp of non-Ty sequence, and a Ty1 element on the Crick strand. In the JSC73-2 strain, the Ty2 element is replaced with KANMX. In the JSC74-1 strain, the space between the two Ty elements, is increased by about 2 kb by insertion of the KANMX marker. In the strain JSC77-1, the centromere-proximal delta element of Ty2 was replaced by the KANMX gene. C. Frequency of recombination between HYG and URA3 in the strain with wild-type HS4 activity and in various deletion derivatives. Asterisks indicate a significant (p<0.05) reduction in recombination activity compared to the control strain.

Figure 7. Model for homologous recombination repair choices depending on the timing of the initiating DSB lesion.

We suggest that most spontaneous DSBs occur during S- or G2 of the cell cycle as indicated by the relative widths of the arrows. The DSBs that occur in G1, however, are repaired preferentially by recombination with the homolog, whereas the DSBs that occur in G2 are repaired preferentially by recombination with the sister chromatids. The rationale for these preferences is given in the text.