Diploid-specific [corrected] genome stability genes of S. cerevisiae: genomic screen reveals haploidization as an escape from persisting DNA rearrangement stress

Abstract

Maintaining a stable genome is one of the most important tasks of every living cell and the mechanisms ensuring it are similar in all of them. The events leading to changes in DNA sequence (mutations) in diploid cells occur one to two orders of magnitude more frequently than in haploid cells. The majority of those events lead to loss of heterozygosity at the mutagenesis marker, thus diploid-specific genome stability mechanisms can be anticipated. In a new global screen for spontaneous loss of function at heterozygous forward mutagenesis marker locus, employing three different mutagenesis markers, we selected genes whose deletion causes genetic instability in diploid Saccharomyces cerevisiae cells. We have found numerous genes connected with DNA replication and repair, remodeling of chromatin, cell cycle control, stress response, and in particular the structural maintenance of chromosome complexes. We have also identified 59 uncharacterized or dubious ORFs, which show the genome instability phenotype when deleted. For one of the strongest mutators revealed in our screen, ctf18Δ/ctf18Δ the genome instability manifests as a tendency to lose the whole set of chromosomes. We postulate that this phenomenon might diminish the devastating effects of DNA rearrangements, thereby increasing the cell's chances of surviving stressful conditions. We believe that numerous new genes implicated in genome maintenance, together with newly discovered phenomenon of ploidy reduction, will help revealing novel molecular processes involved in the genome stability of diploid cells. They also provide the clues in the quest for new therapeutic targets to cure human genome instability-related diseases.

Figure 1. Comparison of genome-wide SLM screen results for CAN1 and URA3 markers.

SLM screen results expressed as averaged LogRatio of relative abundance of each deletion clone obtained for CAN1 and URA3 markers were plotted against each other. LogRatio data derived only from the screens for mutator phenotypes show little correlation (A), whereas after subtracting the LogRatio data expressing resistance to the selection conditions and the LogRatio data expressing growth rate for each deletion strain (B) such a correlation exists.

Figure 2. Example of results of the semi-quantitative SLM drop assay showing various categories of mutator phenotype.

Cell suspensions were serially diluted and spotted onto selection plate (with canavanine or 5′-FOA) and onto dilution control plate as described in Materials and Methods. WT – SLM level in parental strain, M - increased SLM phenotype, HM - high SLM phenotype, ρ⁻ - increased SLM due to respiratory incompetence in WT ρ⁻ strain, Rρ⁻ – resistance to selection conditions acquired along with the loss of respiratory competence, M/GD - high SLM phenotype accompanied by decreased survival rate, seen also without selection, R - full resistance to selection conditions.

Figure 3. DNA content analysis of mutator strains in BY4743 background from homodiploid YKO collection.

DNA content analysis of ctf18Δ/ctf18Δ, ctf8Δ/ctf8Δ, mto1Δ/mto1Δ, phm6Δ/phm6Δ and ted1Δ/ted1Δ strains in BY4743 background from homodiploid YKO collection. Wild-type BY4741 (1n) and BY4743 (2n) strains served as controls for DNA content. Propidium iodide stained cells were analyzed by FACS as described in Materials and Methods.

Figure 4. The changes of DNA content in cells of 2n, 2n ndt80, 2n ctf18 and 2n ndt80 ctf18 strains during prolonged growth.

DNA content analysis was done after: 0, 50, 100, 160, 240 and 320 generations. Please note that “0” represents the starting point of the experiment. In fact, as we estimate, at this point the clones originating from the single zygotes had already grown for about 50 generations. Propidium iodide stained cells were analyzed by FACS as described in Materials and Methods.

Figure 5. The changes of SLM levels in cells of 2n, 2n ndt80, 2n ctf18 and 2n ndt80 ctf18 strains during prolonged growth.

SLM profiles for twenty independent clones of each genotype after growth for the indicated number of generations. SLM profiles for strains: 2n (A), 2n ctf18 (C), 2n ndt80 (E) and 2n ndt80 ctf18 (G) at CAN1 locus. SLM profiles for strains: 2n (B), 2n ctf18 (D), 2n ndt80 (F) and 2n ndt80 ctf18 (H) at URA3 locus. The plots for individual clones are marked with different colors; the plots of the median calculated from the data collected for twenty clones after particular number of generations are indicated by thicker red lines. SLM was measured using semi-quantitative drop assay as described in Materials and Methods.

Figure 6. PFGE analysis of chromosomes from 2n ctf18 clones before and after prolonged growth.

PFGE analysis of chromosomes isolated from eight freshly prepared 2n ctf18 clones (numbered 1 to 8) and from the same clones grown for 240 generations. See Materials and Methods for detailes.

Figure 7. Overrepresentation of GO annotations in the group of 249 genes selected in genomic SLM screen.

The analysis of overrepresentation of Gene Ontology annotations in the group of 249 genes selected in our large scale SLM screen was done with the help of GeneMerge on-line tool (http://genemerge.cbcb.umd.edu/); e<0.1. A) Overrepresentation of Cellular Component annotations. Annotations pertaining to nucleus are shown in green whereas those pertaining to mitochondria are shown in yellow. B) Overrepresentation of Biological Process annotations.