Candida albicans can adapt and grow on sorbose plates by losing one copy of Chr5. Since rad52 mutants of Saccharomyces cerevisiae lose chromosomes at a high rate, we have investigated the ability of C. albicans rad52 to adapt to sorbose. Carad52-DeltaDelta mutants generate Sou(+) strains earlier than wild-type but the final yield is lower, probably because they die at a higher rate in sorbose. As other strains of C. albicans, CAF2 and rad52-DeltaDelta derivatives generate Sou(+) strains by a loss of one copy of Chr5 about 75% of the time. In addition, rad52 strains were able to produce Sou(+) strains by a fragmentation/deletion event in one copy of Chr5, consisting of loss of a region adjacent to the right telomere. Finally, both CAF2 and rad52-DeltaDelta produced Sou(+) strains with two apparent full copies of Chr5, suggesting that additional genomic changes may also regulate adaptation to sorbose.

We have analysed the role of homologous recombination (HR) genes on the repair of double-strand breaks induced by γ-ionising radiation in Candida albicans. Depletion of either CaRad51 or CaRad52 caused a dramatic drop in the number of survivors compared with wild type, whereas depletion of CaRad59 caused a moderate decrease. Besides, compared with Saccharomyces cerevisiae, C. albicans relies more on HR proteins for repair of ionising radiation lesions. Pulse-field electrokaryotypes of survivors identified genetic alterations mainly in the form of aneuploidy in HR mutants and chromosome length polymorphism and ectopic translocation in wild type. Increasing irradiation (4 to 80 krad) of both cycling and nocodazole-treated (G2/M-arrested) cells revealed a gradual loss of chromosomes, larger chromosomes being more affected than smaller ones. For cycling wild-type cells, shattered chromosomes were progressively restored following incubation in yeast extract, peptone, dextrose medium, but not in phosphate-buffered saline, and this accompanied by a moderate increase in colony-forming units, suggesting that repair was followed by replication of survivors. Irradiated G2/M arrested cells from wild type but not from HR mutants partially restored the chromosome ladder following incubation (4-8 hr) in yeast peptone dextrose-nocodazole. However, HR mutants showed a chromosome shattering pattern similar to wild type, an indication that lesions other than double-strand breaks, likely single-strand break, are responsible for their drastically reduced survivability.

No abstract

Candida albicans mutants deficient in homologous recombination (HR) are extremely sensitive to the alkylating agent methyl-methane-sulfonate (MMS). Here, we have investigated the role of HR genes in the protection and repair of C. albicans chromosomes by taking advantage of the heat-labile property (55 °C) of MMS-induced base damage. Acute MMS treatments of cycling cells caused chromosome fragmentation in vitro (55 °C) due to the generation of heat-dependent breaks (HDBs), but not in vivo (30 °C). Following removal of MMS wild type, cells regained the chromosome ladder regardless of whether they were transferred to yeast extract/peptone/dextrose (YPD) or to phosphate buffer saline (PBS); however, repair of HDB/chromosome restitution was faster in YPD, suggesting that it was accelerated by metabolic energy and further fueled by the subsequent overgrowth of survivors. Compared to wild type CAI4, chromosome restitution in YPD was not altered in a Carad59 isogenic derivative, whereas it was significantly delayed in Carad51 and Carad52 counterparts. However, when post-MMS incubation took place in PBS, chromosome restitution in wild type and HR mutants occurred with similar kinetics, suggesting that the exquisite sensitivity of Carad51 and Carad52 mutants to MMS is due to defective fork restart. Overall, our results demonstrate that repair of HDBs by resting cells of C. albicans is rather independent of CaRad51, CaRad52, and CaRad59, suggesting that it occurs mainly by base excision repair (BER).

Candida albicans, the most common fungal pathogen, is a diploid with a genome that is rich in repeats and has high levels of heterozygosity. To study the role of different recombination pathways on direct-repeat recombination, we replaced either allele of the RAD52 gene (Chr6) with the URA-blaster cassette (hisG-URA3-hisG), measured rates of URA3 loss as resistance to 5-fluoroorotic acid (5FOAR) and used CHEF Southern hybridization and SNP-RFLP analysis to identify recombination mechanisms and their frequency in wildtype and recombination mutants. FOAR rates varied little across different strain backgrounds. In contrast, the type and frequency of mechanisms underlying direct repeat recombination varied greatly. For example, wildtype, rad59 and lig4 strains all displayed a bias for URA3 loss via pop-out/deletion vs. inter-homolog recombination and this bias was reduced in rad51 mutants. In addition, in rad51-derived 5FOAR strains direct repeat recombination was associated with ectopic translocation (5%), chromosome loss/truncation (14%) and inter-homolog recombination (6%). In the absence of RAD52, URA3 loss was mostly due to chromosome loss and truncation (80–90%), and the bias of retained allele frequency points to the presence of a recessive lethal allele on Chr6B. However, a few single-strand annealing (SSA)-like events were identified and these were independent of either Rad59 or Lig4. Finally, the specific sizes of Chr6 truncations suggest that the inserted URA-blaster could represent a fragile site.