Open access
Research Article
26 July 2011

Stress Alters Rates and Types of Loss of Heterozygosity in Candida albicans


Genetic diversity is often generated during adaptation to stress, and in eukaryotes some of this diversity is thought to arise via recombination and reassortment of alleles during meiosis. Candida albicans, the most prevalent pathogen of humans, has no known meiotic cycle, and yet it is a heterozygous diploid that undergoes mitotic recombination during somatic growth. It has been shown that clinical isolates as well as strains passaged once through a mammalian host undergo increased levels of recombination. Here, we tested the hypothesis that stress conditions increase rates of mitotic recombination in C. albicans, which is measured as loss of heterozygosity (LOH) at specific loci. We show that LOH rates are elevated during in vitro exposure to oxidative stress, heat stress, and antifungal drugs. In addition, an increase in stress severity correlated well with increased LOH rates. LOH events can arise through local recombination, through homozygosis of longer tracts of chromosome arms, or by whole-chromosome homozygosis. Chromosome arm homozygosis was most prevalent in cultures grown under conventional lab conditions. Importantly, exposure to different stress conditions affected the levels of different types of LOH events, with oxidative stress causing increased recombination, while fluconazole and high temperature caused increases in events involving whole chromosomes. Thus, C. albicans generates increased amounts and different types of genetic diversity in response to a range of stress conditions, a process that we term “stress-induced LOH” that arises either by elevating rates of recombination and/or by increasing rates of chromosome missegregation.
IMPORTANCE Stress-induced mutagenesis fuels the evolution of bacterial pathogens and is mainly driven by genetic changes via mitotic recombination. Little is known about this process in other organisms. Candida albicans, an opportunistic fungal pathogen, causes infections that require adaptation to different host environmental niches. We measured the rates of LOH and the types of LOH events that appeared in the absence and in the presence of physiologically relevant stresses and found that stress causes a significant increase in the rates of LOH and that this increase is proportional to the degree of stress. Furthermore, the types of LOH events that arose differed in a stress-dependent manner, indicating that eukaryotic cells generate increased genetic diversity in response to a range of stress conditions. We propose that this “stress-induced LOH” facilitates the rapid adaptation of C. albicans, which does not undergo meiosis, to changing environments within the host.


Stress-induced mutagenesis fuels the evolution of bacterial pathogens, resistance to antibiotics, tumor progression, and resistance to chemotherapy, all of which occur under stress and are driven by genetic changes (reviewed in references 1 and 2). Although initially controversial because of a lack of distinction between the stress conditions and the conditions used to select for the mutations, the idea that stress-induced mutagenesis induces a hypermutagenic state in subpopulations of bacterial cells is now widely accepted (2). Several studies in Escherichia coli and Saccharomyces cerevisiae focused on genome alterations leading to adaptation in starving or aging cells (3). Importantly, in subsequent starvation episodes, organisms carrying stress-induced genomic rearrangements exhibited a fitness advantage relative to the parental strain (4).
Loss of heterozygosity (LOH) reveals genetic variability in diploid organisms by exposing the phenotypes associated with recessive alleles and often has detrimental outcomes for an organism. For example, LOH is a prerequisite for the initiation and the development of cancer (5) via inactivation of tumor suppressor genes (two-hit hypothesis) (6); indeed, LOH has been associated with breast (7), skin (8), and colorectal (9) cancer. In addition, allele-specific gene expression or allelic imbalance plays a critical biological role in human variability (10). Finally, increased recombination in response to stress (fitness-associated recombination [FAR] [11, 12]) is thought to promote the evolution of complex traits by accelerating the rate of adaptation (13).
LOH events may reveal genetic diversity that affects fitness under stress conditions and/or unstressed conditions. The extent of LOH can range from short LOH tracts that form via gene conversion or double crossovers to long LOH tracts which arise via a single crossover event or by nonreciprocal events such as break-induced replication (BIR), and generate a region of homozygosity extending from the site of recombination to the telomere.
Whole-chromosome (whole-Chr) LOH, a product of chromosome nondisjunction that most frequently occurs because of defects in centromere/kinetochore/mitotic spindle function, usually results in aneuploidy, an imbalance in the number of chromosomes. Most aneuploid strains grow less well than their euploid parental strain. However, under specific stress conditions such as extreme temperature, nutrient shortage, and exposure to chemotherapeutic (14) or antifungal (1517) drugs, some aneuploidies provide a strong fitness advantage. Strains monosomic for a specific chromosome often undergo reduplication of the remaining homolog, yielding a disomic, homozygous chromosome (18). If the two alleles in the heterozygous parental strain provide a differential benefit under a given stress, cells that retain the more beneficial allele after LOH may exhibit a growth advantage over cells that do not undergo LOH.
Candida albicans is a commensal fungus that resides in the human oral cavity, the gastrointestinal tract, and the genitourinary tract. Within the host, it competes with other microbes for nutrition in different body niches and adapts to different temperatures, different pH ranges, and different levels of oxidative stress, for example, when it encounters immune cells such as phagocytes (for a review, see reference 19). Under conditions of weakened immunity or imbalance in the commensal flora, C. albicans becomes an opportunistic pathogen with a disease spectrum ranging from mild superficial infections, such as oral thrush and vaginitis, to severe, life-threatening bloodstream infections such as disseminated candidiasis. These different candidal infections involve colonization of, and thus adaptation to, different host environmental niches and growth conditions, including elevated temperatures in the febrile host. Treatment with antifungal drugs is sometimes accompanied by the rapid acquisition of drug resistance, which can arise by chromosome rearrangements and/or chromosome missegregation (15, 16).
How C. albicans adapts to abrupt changes in environmental conditions is not well understood. C. albicans is a highly heterozygous diploid that reproduces primarily via mitotic division (20, 21). It possesses a mating locus (MAT) (22) and undergoes a parasexual cycle, in which diploids mate (23, 24) to form tetraploids that subsequently undergo “concerted chromosome loss,” producing near-diploid progeny that are often trisomic and/or homozygous for one or more whole chromosomes. Multiple short-range recombination events occur within a subset of parasexual progeny, yielding recombinant chromosomes (25, 26). Homothallic mating further enriches the repertoire of routes that C. albicans can use to generate the wide range of genetic diversity observed in clinical isolates (2729). Importantly, since no meiosis has been detected in C. albicans, genetic diversity is assumed to arise only via mitotic events.
In C. albicans, LOH events, as well as whole-Chr and segmental aneuploidies, have been observed in strains growing under selection for antifungal drug resistance (1517). For example, homozygosis of hyperactive alleles of positive regulators of drug efflux pumps (15, 30) or of genes encoding drug targets, such as ERG11 (31), or increased copy numbers of these genes provide a selective advantage in the presence of these drugs (32). In contrast, growth on poor carbon or nitrogen sources sometimes selects for LOH due to loss of whole chromosomes or chromosome segments (18, 33).
In C. albicans, little is known about the frequency of genetic changes that arise spontaneously and whether the rate of appearance of those changes is affected by exposure to stress. Furthermore, rates of spontaneous LOH have been measured only at one genomic locus (GAL1) (34, 35). Since the types of events that arise at different loci are often influenced by local features of the DNA (e.g., proximity to direct or inverted repeats or transposons can lead to gross chromosomal rearrangements, chromosome aberrations, and overall genome instability in S. cerevisiae [3638]), it is important to determine both the rates and the types of LOH events that occur most frequently and to ask if there are chromosome regions with especially high or low rates of LOH in general. Furthermore, since candidal infections are treated with antifungal drugs and antifungal drug resistance can cause significant clinical complications, it is important to understand how C. albicans cells respond to physiologically relevant stresses, including antifungal drug stress.
Here, we measured rates of LOH and the distribution of types of LOH events that appeared in the absence and in the presence of physiologically relevant stresses. Importantly, we found that stress affected the rates and types of LOH events in a manner that was proportional to the degree of stress. Thus, it appears that C. albicans generates increased genetic diversity in response to a range of stress conditions but that it employs more than one mechanism to do so.


Establishment of a system to measure LOH rates and types in Candida albicans.

We first set out to measure LOH rates at a number of different loci. A critical issue in the measurement of LOH rates is that selection pressure for the marker used should not influence the interpretation of results. We used two counterselectable markers, which greatly facilitate the detection of rare events (39): GAL1 (34, 40), which we were able to use only at its native locus on Chr1, and the URA3 marker, which could be inserted at different loci across the genome. We asked if marker selection affected rates using strain YJB9318, in which one copy of GAL1 was replaced with the URA3 gene. LOH rates calculated for each marker were very similar (see Fig. S1A in the supplemental material), indicating that the two different selection conditions could be used interchangeably and that they did not exert any obvious bias on the LOH rate.
In S. cerevisiae and in limited analyses in C. albicans, LOH is influenced by chromosomal position, including proximity to centromeres (39, 41). We used 18 different strains, each one heterozygous for URA3 at a different locus, with each chromosome arm represented at least once (Fig. 1C; also see the supplemental material). We then compared LOH rates across the genome and as a function of distance from the centromere and other genome features. LOH rates were similar between independent isolates carrying the same marked locus but were different between different marked loci, with LOH rates ranging from ~10−6 to ~10−7/cell division (Fig. 1D; also see Table S1 in the supplemental material). LOH rates at different loci generally correlated with increased distance from the centromere (R2 = 0.3, Fig. S2A) and did not correlate with the distance of the loci from the nearest major repeat sequence (Fig. S2B). The most obvious outlier was the ERG13 locus, which is on the right arm of ChrR adjacent to the rDNA repeats (Fig. 1C; Table S1). This is consistent with the results of Andersen et al. (39), who found nonstereotypical LOH rates for loci near the rDNA repeats in S. cerevisiae. Importantly, the rate of reversion of a point mutation in C. albicans (Fig. 1B; also see the supplemental material) was 5.9 × 10−10 (± 2.1 × 10−10) per generation, which is much lower than rates of LOH measured in this study (see Discussion).
FIG 1 Determination of LOH and point mutation rates. (A) Configuration of Chr1 for comparing LOH rates using selection for loss of GAL1 or URA3. One copy of GAL1 was replaced with URA3. Fluctuation analysis was performed using selection on either 2-deoxygalactose (2-DOG) or 5-fluoroorotic acid (5-FOA). (B) Configuration of Chr4 for measuring the rate of point mutations that revert the his4 G929T point mutation at the native HIS4 locus (18). Fluctuation analysis selected for the appearance of rare His+ colonies. (C) Positions of URA3 insertions (one per strain) used to measure LOH rates. In addition, the position of ADE2, used for half-sector analysis (strain YJB11848, which is ade2∆/∆ at the native ADE2 locus), is shown. (D) LOH rates for markers on all 16 chromosome arms. The distance between the URA3 marker used and the centromere (CEN) is indicated below.
We next determined the types of LOH events using whole-genome single nucleotide polymorphism (SNP) microarray analysis for a set of 21 isolates derived from YJB9318 (GAL1/gal1::URA3) (34) (Fig. 2A). All 21 Gal isolates exhibited alterations on Chr1. Importantly, no SNP changes were detected on other Chrs. Long LOH tracts involving homozygosis of Chr1L, including SNP loci starting between GAL1 and CEN1 and extending to the left telomere (Fig. 2C), were most common. Short LOH tracts surrounding and including the GAL1 locus were less frequent, and only one example of whole-Chr LOH was detected (Fig. 2C).
FIG 2 SNP analysis of LOH types. (A and B) Detection of LOH events using an SNP microarray (A) and using SNP-RFLP analysis (B). (C) Total proportion of each type of LOH event (n = 21 post-fluctuation strains for array analysis [above] and 15 to 24 isolates per marker for 8 different markers yielding n = 173 post-fluctuation analysis strains for SNP-RFLP analysis of the URA3-marked chromosome [below]). (D) Percentages of LOH types from SNP-RFLP analysis at individual Chr arms for the 8 different marked strains tested (n = 15 to 24 for each strain).
Since the SNP microarray analysis described above detected only changes on Chr1 (the marked chromosome), we next analyzed types of LOH events that arose at 8 different loci using 4 SNP-RFLP (restriction fragment length polymorphism) markers located on the same Chr as the URA3 marker that had been lost (42) (Fig. 2B). For each locus studied, we collected 24 5-fluoroorotic acid-resistant (5-FOAr) isolates taken from the 20 independent cultures used to determine LOH rates. When SNP-RFLP markers were distal to the marker that underwent LOH (as they were for URA3 inserts on Chr1, -4, -5, and -6), the analysis distinguished between short LOH tracts (LOH of only URA3 and/or URA3 and the marker closest to it), long LOH tracts (loss of both markers on the same chromosome arm as URA3), and whole-Chr events (LOH of all SNP markers on the chromosome). The SNP-RFLP analysis, like the SNP microarray study, found that long LOH tracts were the most frequent type of LOH event (Fig. 2C; see also Table S2 in the supplemental material).
Long LOH tracts can arise by reciprocal (single crossovers [XO]) or nonreciprocal (BIR) recombination. We used half-sector analysis (see the supplemental material) with a strain carrying a heterozygous ADE2 marker (YJB11848) to distinguish between these two mechanisms (Fig. 3C and D), Interestingly, 88% of the long-tract LOH events arose by a nonreciprocal mechanism such as BIR (see Table S3 in the supplemental material). The remaining 12% arose by a reciprocal recombination event, most likely a single crossover (Fig. 3C).
FIG 3 Half-sector analysis to distinguish reciprocal versus nonreciprocal LOH events. (A) Configuration of Chr5 in strain YJB11848, which has one copy of ADE2 inserted on the left arm of Chr5 and distal SNP marker 10080A. (B) Examples of half-sectored colonies detected on MIN medium supplemented with uridine, histidine, and adenine. (C and D) Reciprocal (C) and nonreciprocal (D) LOH yield half-sectored colonies in which the phenotype of the white sector is homozygous or heterozygous, respectively. Adapted from the work of Andersen et al. (39).

Stress conditions cause elevated rates of LOH and aneuploidy.

LOH and aneuploidy are found frequently in strains resistant to azole antifungal drugs (17, 43) and can arise multiple times in an individual patient (32). Furthermore, at least some aneuploidies clearly confer a fitness advantage over the parental strain in the presence of azoles and, perhaps surprisingly, exert little, if any, fitness cost in the absence of the drugs (16). In addition, in S. cerevisiae, while point mutations can increase the fitness of aneuploid strains (44), different combinations of aneuploid chromosomes also can confer a fitness advantage under different types of stress conditions (14).
We asked if physiologically relevant stress conditions that C. albicans regularly encounters in the host (increased temperature [39°C, mimicking moderate fever in the host], oxidative stress [hydrogen peroxide, mimicking the production of reactive oxygen species by the host’s immune cells], or antifungal stress [fluconazole, the most widely and commonly used antifungal drug, at subinhibitory concentration]) affected the rate at which LOH and/or aneuploidy arises. Importantly, because stress conditions can be selective for specific aneuploidies or LOH events, these analyses were performed using six different parental strains, three strains with URA3 inserted at different positions on Chr5 and three strains with URA3 inserted on Chr1, -2, or -7 (Fig. 1C; location of URA3 is indicated by brown boxes with black frame; see also Table S6 in the supplemental material). Chr 5 has been the focus of several studies on genome instability (15, 17, 45), and Chr1 (~3.2 Mbp), Chr2 (~2.2 Mbp), and Chr7 (~0.95 Mbp) were chosen to account for differences in Chr size that potentially could influence LOH rates and events. Furthermore, by measuring LOH rates using several different strains with markers on different chromosomes, there is less likelihood that selection pressure on a given gene or chromosome region will affect the general conclusions.
A critical point is that the URA3 marker does not have a known role in the in vitro survival of C. albicans cells under heat, oxidative, or antifungal stress. In addition, we performed the stress experiments with strain YJB9318, which has URA3 inserted at the GAL1 locus, and strains were analyzed as described above but with selection for GAL1 loss (on 2-deoxygalactose [2-DOG]) or URA3 loss (on 5-FOA) after exposure to the stress condition. Importantly, similar to the results under nonstress conditions, LOH rates of the two markers under stress conditions were not different (see Fig. S1A and Table S4 in the supplemental material), and thus any change in LOH rate is independent of the counterselectable marker used and can be attributed to the particular stress condition.
Strikingly, LOH rates were higher under stress conditions than under unstressed conditions. Fold changes for LOH rates were variable and yet lower on average for temperature stress (ranging from a 1- to a 40-fold increase) and intermediate for oxidative stress (from a 3- to a 72-fold increase). Fluconazole exposure resulted in the most dramatic increases in LOH rate, with up to a 285-fold increase (Fig. 4; see also Table S5 in the supplemental material). While strain YJB9834 appeared to be an outlier with respect to LOH rates in all three stresses (Table S5 and Fig. 4), these differences were not statistically significant (Kruskal-Wallis; P = 0.45 for each stress). Additional statistical analyses were performed for each stress to determine whether LOH rates are significantly different between control and stress conditions. While there was no statistically significant difference observed for temperature stress (U = 54; P = 0.125), for both H2O2 and fluconazole stress control LOH rates were statistically different from the stress LOH rates (U = 0; P < 0.001).
FIG 4 LOH rate fold changes in stressed cells. The graph shows fold increases in LOH rates in stressed cells for 6 different strains each with URA3 inserted at a different locus (Fig. 1C; see also Table S6 in the supplemental material). The y axis shows fold change in LOH rate (note that the y axis is on a logarithmic scale).

Growth under stress conditions alters the proportion of different LOH events that are recovered.

We next analyzed the types of LOH events that accompanied the loss of URA3 during exposure to stress. For each of the three stress conditions, we analyzed 24 5-FOAr isolates derived from each of the strains carrying a marked locus, using SNP-RFLP analysis. Interestingly, the proportions of short LOH tracts, long LOH tracts, and whole-Chr LOH differed depending upon the stress condition. In strains exposed to 39°C, a stress that did not cause a statistically significant increase in LOH rate (see above), the proportion of whole-Chr LOH increased ~6-fold (Fig. 5A; see also Table S2 in the supplemental material). In contrast, in cells exposed to H2O2, the proportion of short LOH tracts increased almost 2-fold (Fig. 5A; Table S2). Finally, exposure to fluconazole, which caused the most dramatic increase in LOH rate, resulted in a ~5-fold increase in the proportion of whole-Chr LOH (Fig. 5A). The rate of LOH did not correlate with the type of LOH events seen: both febrile temperature (no increase in LOH rate) and an azole antifungal (285-fold increase in LOH rate) resulted in an increased proportion of whole-Chr LOH. Furthermore, exposure to different stresses triggered different mechanisms of genome alteration: oxidative stress caused an increased proportion of short LOH tracts, while the other two stresses caused an increased proportion of whole-Chr LOH, presumably due to elevated chromosome nondisjunction.
FIG 5 Effect of stress on types and rates of LOH events. (A) Short-tract, long-tract, and whole-Chr events are diagrammed above and framed with a color corresponding to the type: cyan for short-tract, light gray for long-tract, and lavender for whole-Chr events. (B) LOH rates in cells exposed to different concentrations of H2O2 (R2 = 0.96). (C) Types of LOH events in cells exposed to different H2O2 concentrations shift from primarily short-tract events to primarily long-tract events. Color scheme is as in panel A.

Increased H2O2 stress causes increased LOH rates and is accompanied by a shift in LOH types.

The effects of stress on LOH rates in the experiments above appeared to be proportional to the degree to which cells were stressed (with H2O2 < temperature < fluconazole, measured as reduction in growth rates and levels of viability [see Fig. S1B and S2C to S2F in the supplemental material]), with the three types of stresses having very different physiological consequences. Thus, we next asked if LOH rates were directly related to the degree of stress using a single type of stress. To determine the degree to which H2O2 stress affected cell growth, we first performed growth curves and fluctuation analysis using a range of H2O2 concentrations. YJB9318 grew as well as did the no-stress control at up to 1 mM H2O2 (Fig. S1B), which suggests that the 0.4 mM stress condition used above (and in several other C. albicans studies [46, 47]) should be considered a mild stress condition. Growth decreased at 3 mM H2O2 and stopped completely at 10 mM H2O2. Fluctuation analysis was performed using the same H2O2 concentrations as those for the growth curves, and FUN-1 staining was performed to correct for nonviable cells. Percent dead cells ranged from 0.6% in the no-stress controls to 7% in 10 mM H2O2 (data not shown).
Importantly, the rate of LOH increased linearly with the increased level of H2O2 used (R2 = 0.9) (Fig. 5B). This linear relationship was also observed for different temperatures and a range of fluconazole concentrations (see Fig. S2C to S2F in the supplemental material). In addition, SNP-RFLP analysis was performed for 24 5-FOAr isolates after fluctuation analysis for 1 mM, 5 mM, and 10 mM H2O2 stress to determine the LOH events accompanying URA3 loss. Strikingly, SNP-RFLP analysis revealed a shift in LOH types from 100% short LOH tracts for no stress and 0.4 mM H2O2 stress to mostly long LOH tracts for 1 mM, 5 mM, and 10 mM H2O2 stress (Fig. 5C). Since the SNP-RFLP analysis followed only two markers per Chr arm, we cannot rule out the possibility that long LOH tracts were due to two separate short LOH events. The most parsimonious explanation is that the large increase in the proportion of these events is due to increases in single crossovers and/or BIR events. Perhaps, the proportion of double-strand breaks (DSBs) relative to single-strand nicks increases with increasing H2O2 concentration. Thus, for a single type of stress, it is clear that increasing levels of stress result in increased LOH rates and a shift in the types of events that occur.


In asexual organisms, mitotic recombination is a major mechanism of genome evolution (48, 49). Here, we exploited the naturally high levels of heterozygosity in C. albicans to measure LOH events that reveal genetic diversity at the population level. Importantly, we found that LOH rates increased with increased levels of stress and that different stress conditions caused increases in different types of LOH events, many of which have been seen in clinical isolates (reviewed in reference 50) as well as lab isolates that were passaged through a mammalian host (34). Increasing rates of LOH in the presence of stress may facilitate the rapid adaptation of C. albicans populations to changing environments within the host, and the types of LOH events that arise likely reflect the stress conditions to which the population was exposed.

Rates and types of LOH in C. albicans.

LOH rates measured here for C. albicans are similar to those reported previously for S. cerevisiae (39, 51). Similarly, the rate of reversion of a point mutation in C. albicans was comparable to rates of point mutations in S. cerevisiae, Neurospora crassa, and E. coli (51, 52). Importantly, C. albicans LOH rates are several orders of magnitude more frequent than point mutation rates. While we do not have strains that allow us to directly compare point mutation and LOH rates at the same locus, the data are consistent with the idea that LOH events are much more frequent than point mutations.
The types of LOH events that arise in C. albicans are independent of the marker employed or the method used to identify LOH events (Fig. 2A to 2C) and yet are locus dependent both in the absence and in the presence of stress (Fig. 2D). Similarly, in S. cerevisiae, locus-specific effects on the types of LOH have been observed (39). Thus, despite the lack of meiosis, C. albicans undergoes a range of recombination events that resemble those seen in an organism with a conventional sexual cycle.
It is not clear if whole-Chr homozygosis or long LOH tracts occur more efficiently or whether they provide more benefit (or less fitness cost) during growth under stress conditions. For example, chromosome nondisjunction, which causes whole-Chr LOH, may be less costly to the cell than activation of the recombination machinery, which is involved in generating short and long LOH tracts. In addition, whole-chromosome events may provide a larger opportunity to provide a selective advantage and yet can also incur a larger fitness cost (14, 44, 53, 54).

The influence of stress on rates and types of LOH events.

Different stress conditions clearly increase the levels of different types of LOH. We propose that this is directly related to the type of DNA damage inflicted by the stress condition. Temperature stress causes chromosome loss (55), as well as aneuploidy (56), and we speculate that this occurs through a limitation in heat shock protein chaperones and cochaperones Sgt1, Sti1, and Cdc37, which have roles in the assembly of functional kinetochores (reviewed in reference 57) and spindle pole bodies/centrosomes (58) and the function of mitotic checkpoints (59).
The mechanism by which fluconazole treatment causes whole-Chr LOH is less obvious. However, fluconazole-resistant (Flur) strains are often aneuploid (16, 17), consistent with the idea that fluconazole affects chromosome segregation, either directly or indirectly. For example, we imagine that alterations in ergosterol biosynthesis may affect nuclear membrane fluidity, which would then impact the function of the spindle pole bodies embedded in the nuclear membrane (reviewed in reference 60) and/or the ability of cytoplasmic dynein to properly associate with the cortex and execute proper nuclear separation (61). Such defects would indirectly affect chromosome segregation mechanisms, resulting in polyploidy and/or aneuploidy.
All organisms are constantly exposed to oxidative stress as a normal by-product of respiratory metabolism (62, 63). In S. cerevisiae, oxidative stress caused an increase in double-strand breaks (DSBs) (64, 65) and a ~10-fold increase in mitotic recombination rates (66). Similarly, H2O2 caused a dose-dependent increase in mutation frequency at the human leukocyte antigen class A locus in human T lymphocytes (65). We propose that H2O2 in C. albicans causes increased LOH rates by generating chromosome breaks that are repaired by recombination, resulting in the linear relationship between LOH rate and H2O2 concentration.

Benefits and costs of LOH.

The C. albicans genome and population structure suggests that the organism is generally clonal (20, 21) and, as we show here, gives rise to long LOH tracts during mitotic growth (Fig. 2D and 5A and C; see also Table S2 in the supplemental material). Interestingly, a recent study revealed that the distribution of genetic variation in clinical isolates of S. cerevisiae differs from that of environmental isolates (67). Like C. albicans, clinical S. cerevisiae isolates exhibited high levels of heterozygosity, reduced sporulation efficiency, and more efficient pseudohyphal formation compared to environmental isolates (67). Strikingly, heterozygous clinical S. cerevisiae isolates had long LOH tracts, consistent with long periods of clonality predicted by the facultative asexuality hypothesis (68), where diploid asexual organisms persist by means of mitotic recombination. Evidence for long-tract LOH events in the evolutionary past also can be found in C. albicans strain SC5314 (69), as well as in strain WO-1 (69). Thus, it appears that, as in S. cerevisiae, the stress of growth within host niches has had an important influence on the genome structure of C. albicans.
Relevant to this work, C. albicans and other pathogenic Candida species can adapt to very high levels of oxidative stress (70, 71) while S. cerevisiae laboratory strains cannot. It would be interesting to determine if clinical S. cerevisiae isolates, which have LOH patterns more like those seen in C. albicans, also adapt to oxidative stress more readily, since adaptation to oxidative stress is thought to be important for survival of an initial oxidative attack, as well as for the subsequent establishment of an infection.
While homozygosis of some alleles may provide a fitness advantage under stress and may not have a major fitness cost as measured under laboratory conditions (16, 72, 73), the fact that clinical C. albicans strains carry homozygous regions, and yet heterozygosity remains prevalent at most loci, implies that there is an advantage to the maintenance of heterozygosity during growth in the human host. Whether cells that undergo LOH are eventually lost from the population or whether they regain heterozygosity through rare mating events (74) remains to be determined.


Determination of LOH rates by fluctuation analysis.

To measure LOH rates, fluctuation analysis was essentially carried out as described previously (56). Briefly, strains were streaked for single colonies on either MIN (0.67% yeast nitrogen base without amino acids, 2% dextrose, 1.5% agar) (5-FOA selection) or MIN-Gal (2-DOG selection) with appropriate amino acids and grown for 2 days at 30°C. For nonstress conditions, 20 independent single colonies each were inoculated into 5-ml liquid yeast extract-peptone-dextrose (YPD; 1% yeast extract, 1% Bacto peptone, 2% dextrose, 1.5% agar) cultures and grown for 16 h at 30°C. After 16 h of growth, cells were pelleted, washed once with distilled water, and resuspended in 1 ml distilled water. Appropriate dilutions of each culture were spotted onto YPD for total cell count and onto 5-FOA to determine the proportion of cells that lost the URA3 marker (5-FOAr) or onto 2-DOG to determine the proportion of Gal cells (2-DOGr). YPD CFU were counted on day 2, and 5-FOAr colonies were counted on day 3, which is prior to the time that colonies would have formed “adaptive” mutations on the 5-FOA medium (75). URA3 and GAL1 loss rates were determined using the method of the median by Lea and Coulson (76).
Standard deviations were calculated in Excel. A Kruskal-Wallis test was performed to test if LOH rates at the different genomic loci were statistically different from each other. For each Chr arm locus, two independent transformants were analyzed (see Table S6 in the supplemental material) and at least two independent fluctuation analyses were performed for each transformant.

Determination of LOH rates under stress.

To determine whether stress alters LOH rates, strains were grown using three different stress conditions that mimic different host environments: increased temperature mimicking a mild fever, oxidative stress (hydrogen peroxide [H2O2]) mimicking C. albicans interaction with host immune cells, and antifungal drug (fluconazole) stress. Twenty cultures, each inoculated with a single colony, were grown at 30°C in YPD containing 0.4 mM H2O2 or 1 µg/ml fluconazole for 16 h. For temperature stress, the cultures were grown in YPD at 39°C for 16 h. In parallel, for each stress experiment, 20 cultures were grown at 30°C in YPD and served as no-stress controls. Fluctuation analysis was carried out at least twice for each strain under each condition. LOH rates were determined as described above. To determine if an increase in severity of a particular stress resulted in increased LOH rates, fluctuation analysis was performed using a range of H2O2 concentrations (0.4 mM, 0.6 mM, 0.8 mM, 1 mM, 3 mM, 5 mM, 7 mM, and 10 mM in strain YJB9318); different temperatures for strains YJB9318, YJB9984, and YJB9834; and a range of fluconazole concentrations (0 µg/ml, 0.4 µg/ml, 0.8 µg/ml, 1.0 µg/ml, and 1.4 µg/ml for strain YJB9318). LOH rates were measured as described above, and determination of doubling times, growth curves, and FUN-1 staining are described in the supplemental material. A nonparametric Mann-Whitney U test was performed for each stress to test if LOH rates for stressed conditions were statistically different from those for unstressed conditions.

Determination of LOH events.

SNP microarray analysis was performed as described elsewhere (26). We previously developed an SNP-RFLP assay for distinguishing short-tract LOH, long-tract LOH, and whole-Chr LOH events in C. albicans using a set of 32 SNP-RFLP markers, 4 per Chr and 2 markers per Chr arm (42). To distinguish the LOH events that accompanied the loss of the URA3 marker, for Chr1, -4, -5, and -6, 24 5-FOAr isolates obtained after fluctuation analysis were analyzed by SNP-RFLP for the chromosome carrying the URA3 marker (42). Briefly, PCR was performed in 96-well plate format in a total volume of 25 µl with 5× PCR buffer; 2.5 mM (each) dATP, dCTP, dGTP, and dTTP; 10 µM (each) primer (see Table S7 in the supplemental material); 0.25 µl e2TAK (Takara); and 30 ng of genomic DNA (gDNA) using the following conditions: initial denaturation at 98°C for 3 min; 30 cycles with denaturation at 98°C for 10 s, annealing at 55°C for 10 s, and extension at 72°C for 1 min; and a final extension time at 72°C for 5 min. For restriction digests, 5 µl of PCR product was digested overnight in a total volume of 15 µl with 1 µl restriction enzyme, 10× restriction buffer, and 10× bovine serum albumin (where indicated) at 37°C or 65°C (for details, see Table S7 in reference 42). Digested PCR products were separated in a 3% agarose gel along with undigested and digested controls, gels were photographed, and restriction patterns were analyzed.
For details of materials and methods not given here, please see Text S1 in the supplemental material.


The project described was supported by NIAID R01AI0624273 to J.B. and by award no. P20RR016463 (INBRE) from the National Center for Research Resources to T.P. M.A.W. and D.B. were supported by UROP funding.
We thank Karen Tang for producing the Excel script to calculate LOH rates; Suzanne Lovett for help with statistical analyses; Maryam Gerami-Nejad for strain construction; and Mark McClellan, Merima Helic, and Ari Holloway-Nahum for technical assistance. For helpful discussions and comments on the manuscript, we are grateful to R. Li, D. Gottschling, A. Selmecki, M. Hickman, and M. Anderson.

Supplemental Material

File (mbo004111142s1.doc)
File (mbo004111142sf01.pdf)
File (mbo004111142sf02.pdf)
File (mbo004111142st1.doc)
File (mbo004111142st2.doc)
File (mbo004111142st3.doc)
File (mbo004111142st4.doc)
File (mbo004111142st5.doc)
File (mbo004111142st6.doc)
File (mbo004111142st7.doc)
ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.


Drake JW. 2007. Too many mutants with multiple mutations. Crit. Rev. Biochem. Mol. Biol. 42:247–258.
Galhardo RS, Hastings PJ, and Rosenberg SM. 2007. Mutation as a stress response and the regulation of evolvability. Crit. Rev. Biochem. Mol. Biol. 42:399–435.
McMurray MA and Gottschling DE. 2003. An age-induced switch to a hyper-recombinational state. Science 301:1908–1911.
Coyle S and Kroll E. 2008. Starvation induces genomic rearrangements and starvation-resilient phenotypes in yeast. Mol. Biol. Evol. 25:310–318.
Lengauer C, Kinzler KW, and Vogelstein B. 1998. Genetic instabilities in human cancers. Nature 396:643–649.
Knudson AG. 1993. Antioncogenes and human cancer. Proc. Natl. Acad. Sci. U. S. A. 90:10914–10921.
Chen X et al. 2008. Allelic imbalance in BRCA1 and BRCA2 gene expression is associated with an increased breast cancer risk. Hum. Mol. Genet. 17:1336–1348.
Purdie KJ et al. 2007. Allelic imbalances and microdeletions affecting the PTPRD gene in cutaneous squamous cell carcinomas detected using single nucleotide polymorphism microarray analysis. Genes Chromosomes Cancer 46:661–669.
Yan H et al. 2002. Small changes in expression affect predisposition to tumorigenesis. Nat. Genet. 30:25–26.
Yan H, Yuan W, Velculescu VE, Vogelstein B, and Kinzler KW. 2002. Allelic variation in human gene expression. Science 297:1143.
Hadany L and Beker T. 2003. Fitness-associated recombination on rugged adaptive landscapes. J. Evol. Biol. 16:862–870.
Schoustra S, Rundle HD, Dali R, and Kassen R. 2010. Fitness-associated sexual reproduction in a filamentous fungus. Curr. Biol. 20:1350–1355.
Hadany L and Beker T. 2003. On the evolutionary advantage of fitness-associated recombination. Genetics 165:2167–2179.
Pavelka N et al. 2010. Aneuploidy confers quantitative proteome changes and phenotypic variation in budding yeast. Nature 468:321–325.
Coste A et al. 2006. A mutation in Tac1p, a transcription factor regulating CDR1 and CDR2, is coupled with loss of heterozygosity at chromosome 5 to mediate antifungal resistance in Candida albicans. Genetics 172:2139–2156.
Selmecki A, Dulmage K, Cowen L, Anderson JB, and Berman J. 2009. Aneuploidy provides increased fitness during the evolution of antifungal drug resistance. PLoS Genet. 5:e1000705.
Selmecki A, Forche A, and Berman J. 2006. Aneuploidy and isochromosome formation in drug-resistant Candida albicans. Science 313:367–370.
Janbon G, Sherman F, and Rustchenko E. 1998. Monosomy of a specific chromosome determines L-sorbose utilization: a novel regulatory mechanism in Candida albicans. Proc. Natl. Acad. Sci. U. S. A. 95:5150–5155.
Calderone R. 2002. Candida and candidiasis. American Society for Microbiology, Washington, DC.
Forche A, Schönian G, Gräser Y, Vilgalys R, and Mitchell TG. 1999. Genetic structure of typical and atypical populations of Candida albicans from Africa. Fungal Genet. Biol. 28:107–125.
Gräser Y et al. 1996. Molecular markers reveal that population structure of the human pathogen Candida albicans exhibits both clonality and recombination. Proc. Natl. Acad. Sci. U. S. A. 93:12473–12477.
Hull CM and Johnson AD. 1999. Identification of a mating type-like locus in the asexual pathogenic yeast Candida albicans. Science 285:1271–1275.
Hull CM, Raisner RM, and Johnson AD. 2000. Evidence for mating of the “asexual” yeast Candida albicans in a mammalian host. Science 289:307–310.
Magee BB and Magee PT. 2000. Induction of mating in Candida albicans by construction of MTLa and MTLalpha strains. Science 289:310–313.
Bennett RJ and Johnson AD. 2003. Completion of a parasexual cycle in Candida albicans by induced chromosome loss in tetraploid strains. EMBO J. 22:2505–2515.
Forche A et al. 2008. The parasexual cycle in Candida albicans provides an alternative pathway to meiosis for the formation of recombinant strains. PLoS Biol. 6:e110.
Boerlin P et al. 1996. Typing Candida albicans oral isolates from human immunodeficiency virus-infected patients by multilocus enzyme electrophoresis and DNA fingerprinting. J. Clin. Microbiol. 34:1235–1248.
Schmid J, Voss E, and Soll DR. 1990. Computer-assisted methods for assessing strain relatedness in Candida albicans by fingerprinting with the moderately repetitive sequence Ca3. J. Clin. Microbiol. 28:1236–1243.
Tavanti A et al. 2005. Population structure and properties of Candida albicans, as determined by multilocus sequence typing. J. Clin. Microbiol. 43:5601–5613.
Dunkel N, Blass J, Rogers PD, and Morschhäuser J. 2008. Mutations in the multi-drug resistance regulator MRR1, followed by loss of heterozygosity, are the main cause of MDR1 overexpression in fluconazole-resistant Candida albicans strains. Mol. Microbiol. 69:827–840.
Franz R et al. 1998. Multiple molecular mechanisms contribute to a stepwise development of fluconazole resistance in clinical Candida albicans strains. Antimicrob. Agents Chemother. 42:3065–3072.
Selmecki AM, Gerami-Nejad M, Paulson C, Forche A, and Berman J. 2008. An isochromosome confers drug resistance in vivo by amplification of two genes, ERG11 and TAC1. Mol. Microbiol. 68:624–641.
Rustchenko EP, Howard DH, and Sherman F. 1994. Chromosomal alterations of Candida albicans are associated with the gain and loss of assimilating functions. J. Bacteriol. 176:3231–3241.
Forche A, Magee PT, Selmecki A, Berman J, and May G. 2009. Evolution in Candida albicans populations during single passage through a mouse host. Genetics 182:799–811.
Legrand M et al. 2008. Haplotype mapping of a diploid non-meiotic organism using existing and induced aneuploidies. PLoS Genet. 4:e1.
Argueso JL et al. 2008. Double-strand breaks associated with repetitive DNA can reshape the genome. Proc. Natl. Acad. Sci. U. S. A. 105:11845–11850.
Paek AL et al. 2009. Fusion of nearby inverted repeats by a replication-based mechanism leads to formation of dicentric and acentric chromosomes that cause genome instability in budding yeast. Genes Dev. 23:2861–2875.
Tang W et al. 2011. Friedreich’s ataxia (GAA)/(TTC) repeats strongly stimulate mitotic crossovers in Saccharomyces cerevisae. PLoS Genet. 7:e1001270.
Andersen MP, Nelson ZW, Hetrick ED, and Gottschling DE. 2008. A genetic screen for increased loss of heterozygosity in Saccharomyces cerevisiae. Genetics 179:1179–1195.
Legrand M, Chan CL, Jauert PA, and Kirkpatrick DT. 2007. Role of DNA mismatch repair and double-strand break repair in genome stability and antifungal drug resistance in Candida albicans. Eukaryot. Cell 6:2194–2205.
Lephart PR, Chibana H, and Magee PT. 2005. Effect of the major repeat sequence on chromosome loss in Candida albicans. Eukaryot. Cell 4:733–741.
Forche A, Steinbach M, and Berman J. 2009. Efficient and rapid identification of Candida albicans allelic status using SNP-RFLP. FEMS Yeast Res. 9:1061–1069.
Coste A et al. 2007. Genotypic evolution of azole resistance mechanisms in sequential Candida albicans isolates. Eukaryot. Cell 6:1889–1904.
Torres EM et al. 2010. Identification of aneuploidy-tolerating mutations. Cell 143:71–83.
Wu W, Lockhart SR, Pujol C, Srikantha T, and Soll DR. 2007. Heterozygosity of genes on the sex chromosome regulates Candida albicans virulence. Mol. Microbiol. 64:1587–1604.
Alvarez-Peral FJ, Zaragoza O, Pedreno Y, and Argüelles J-C. 2002. Protective role of trehalose during severe oxidative stress caused by hydrogen peroxide and the adaptive oxidative stress response in Candida albicans. Microbiology 148:2599–2606.
Enjalbert B, Nantel A, and Whiteway M. 2003. Stress-induced gene expression in Candida albicans: absence of a general stress response. Mol. Biol. Cell 14:1460–1467.
Barbera MA and Petes TD. 2006. Selection and analysis of spontaneous reciprocal mitotic cross-overs in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. U. S. A. 103:12819–12824.
Symington LS. 2002. Role of RAD52 epistasis group genes in homologous recombination and double-strand break repair. Microbiol. Mol. Biol. Rev. 66:630–670.
Rustchenko E. 2007. Chromosome instability in Candida albicans. FEMS Yeast Res. 7:2–11.
Drake JW. 1999. The distribution of rates of spontaneous mutation over viruses, prokaryotes, and eukaryotes. Ann. N. Y. Acad. Sci. 870:100–107.
Drake JW. 1991. A constant rate of spontaneous mutation in DNA-based microbes. Proc. Natl. Acad. Sci. U. S. A. 88:7160–7164.
Pavelka N, Rancati G, and Li R. 2010. Dr Jekyll and Mr Hyde: role of aneuploidy in cellular adaptation and cancer. Curr. Opin. Cell Biol. 22:809–815.
Rancati G et al. 2008. Aneuploidy underlies rapid adaptive evolution of yeast cells deprived of a conserved cytokinesis motor. Cell 135:879–893.
Hilton C, Markie D, Corner B, Rikkerink E, and Poulter R. 1985. Heat shock induces chromosome loss in the yeast Candida albicans. Mol. Gen. Genet. 200:162–168.
Bouchonville K, Forche A, Tang KE, Selmecki A, and Berman J. 2009. Aneuploid chromosomes are highly unstable during DNA transformation of Candida albicans. Eukaryot. Cell 8:1554–1566.
Pearl LH and Prodromou C. 2006. Structure and mechanism of the Hsp90 molecular chaperone machinery. Annu. Rev. Biochem. 75:271–294.
Lange BMH, Bachi A, Wilm M, and González C. 2000. Hsp90 is a core centrosomal component and is required at different stages of the centrosome cycle in Drosophila and vertebrates. EMBO J. 19:1252–1262.
Felts SJ, Karnitz LM, and Toft DO. 2007. Functioning of the Hsp90 machine in chaperoning checkpoint kinase 1 (Chk1) and the progesterone receptor (PR). Cell Stress Chaperones 12:353–363.
Webster M, Witkin KL, and Cohen-Fix O. 2009. Sizing up the nucleus: nuclear shape, size and nuclear-envelope assembly. J. Cell Sci. 122:1477–1486.
Finley KR, Bouchonville KJ, Quick A, and Berman J. 2008. Dynein-dependent nuclear dynamics affect morphogenesis in Candida albicans by means of the Bub2p spindle checkpoint. J. Cell Sci. 121:466–476.
Finkel T and Holbrook NJ. 2000. Oxidants, oxidative stress and the biology of ageing. Nature 408:239–247.
Fridovich I. 1995. Superoxide radical and superoxide dismutases. Annu. Rev. Biochem. 64:97–112.
Steinboeck F et al. 2010. The relevance of oxidative stress and cytotoxic DNA lesions for spontaneous mutagenesis in non-replicating yeast cells. Mutat. Res. 688:47–52.
Turner MJ and Slack FJ. 2009. Transcriptional control of microRNA expression in C. elegans: promoting better understanding. RNA Biol. 6:49–53.
Jamieson DJ. 1998. Oxidative stress responses of the yeast Saccharomyces cerevisiae. Yeast 14:1511–1527.
Magwene PM et al. 2011. Outcrossing, mitotic recombination, and life-history trade-offs shape genome evolution in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. U. S. A. 108:1987–1992.
Mandegar MA and Otto SP. 2007. Mitotic recombination counteracts the benefits of genetic segregation. Proc. Biol. Sci. 274:1301–1307.
Butler G et al. 2009. Evolution of pathogenicity and sexual reproduction in eight Candida genomes. Nature 459:657–662.
González-Párraga P, Hernández JA, and Argüelles JC. 2003. Role of antioxidant enzymatic defences against oxidative stress (H2O2) and the acquisition of oxidative tolerance in Candida albicans. Yeast 20:1161–1169.
Jamieson DJ, Stephen DW, and Terriere EC. 1996. Analysis of the adaptive oxidative stress response of Candida albicans. FEMS Microbiol. Lett. 138:83–88.
Cowen LE, Kohn LM, and Anderson JB. 2001. Divergence in fitness and evolution of drug resistance in experimental populations of Candida albicans. J. Bacteriol. 183:2971–2978.
Zeyl C and DeVisser JAGM. 2001. Estimates of the rate and distribution of fitness effects of spontaneous mutation in Saccharomyces cerevisiae. Genetics 157:53–61.
de Clare M, Pir P, and Oliver SG. 2011. Haploinsufficiency and the sex chromosomes from yeasts to humans. BMC Biol. 9:15.
Rosenberg SM and Hastings PJ. 2004. Adaptive point mutation and adaptive amplification pathways in the Escherichia coli Lac system: stress responses producing genetic change. J. Bacteriol. 186:4838–4843.
Lea DE and Coulson CA. 1949. The distribution of the numbers of mutants in bacterial populations. J. Genet. 49:264–285.

Information & Contributors


Published In

cover image mBio
Volume 2Number 41 September 2011
eLocator: 10.1128/mbio.00129-11
Editor: John C. Boothroyd, Stanford University


Received: 20 June 2011
Accepted: 30 June 2011
Published online: 26 July 2011



A. Forche
Department of Biology, Bowdoin College, Brunswick, Maine, USA
D. Abbey
Department of Genetics, Cell Biology and Development, University of Minnesota, Minneapolis, Minnesota, USA
T. Pisithkul
Department of Biology, Bowdoin College, Brunswick, Maine, USA
M. A. Weinzierl
Department of Genetics, Cell Biology and Development, University of Minnesota, Minneapolis, Minnesota, USA
T. Ringstrom
Department of Genetics, Cell Biology and Development, University of Minnesota, Minneapolis, Minnesota, USA
D. Bruck
Department of Genetics, Cell Biology and Development, University of Minnesota, Minneapolis, Minnesota, USA
K. Petersen
Department of Genetics, Cell Biology and Development, University of Minnesota, Minneapolis, Minnesota, USA
J. Berman
Department of Genetics, Cell Biology and Development, University of Minnesota, Minneapolis, Minnesota, USA


John C. Boothroyd
Stanford University


Address correspondence to J. Berman, [email protected].

Metrics & Citations


Note: There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.

Citation counts come from the Crossref Cited by service.


If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.

View Options

Figures and Media






Share the article link

Share with email

Email a colleague

Share on social media

American Society for Microbiology ("ASM") is committed to maintaining your confidence and trust with respect to the information we collect from you on websites owned and operated by ASM ("ASM Web Sites") and other sources. This Privacy Policy sets forth the information we collect about you, how we use this information and the choices you have about how we use such information.
FIND OUT MORE about the privacy policy