Skip to main content

Some NLM-NCBI services and products are experiencing heavy traffic, which may affect performance and availability. We apologize for the inconvenience and appreciate your patience. For assistance, please contact our Help Desk at info@ncbi.nlm.nih.gov.

eLife logoLink to eLife
. 2020 Jul 20;9:e58349. doi: 10.7554/eLife.58349

Expandable and reversible copy number amplification drives rapid adaptation to antifungal drugs

Robert T Todd 1, Anna Selmecki 1,
Editors: Kevin J Verstrepen2, Patricia J Wittkopp3
PMCID: PMC7371428  PMID: 32687060

Abstract

Previously, we identified long repeat sequences that are frequently associated with genome rearrangements, including copy number variation (CNV), in many diverse isolates of the human fungal pathogen Candida albicans (Todd et al., 2019). Here, we describe the rapid acquisition of novel, high copy number CNVs during adaptation to azole antifungal drugs. Single-cell karyotype analysis indicates that these CNVs appear to arise via a dicentric chromosome intermediate and breakage-fusion-bridge cycles that are repaired using multiple distinct long inverted repeat sequences. Subsequent removal of the antifungal drug can lead to a dramatic loss of the CNV and reversion to the progenitor genotype and drug susceptibility phenotype. These findings support a novel mechanism for the rapid acquisition of antifungal drug resistance and provide genomic evidence for the heterogeneity frequently observed in clinical settings.

Research organism: Other

Introduction

The evolution of antifungal drug resistance is an urgent threat to human health worldwide, particularly for hospitalized and immune-compromised individuals (Perea and Patterson, 2002; Pfaller, 2012; Vandeputte et al., 2012). Only three classes of antifungal drugs are currently available and resistance to all three classes occurred for the first time in the emerging fungal pathogen Candida auris (Chen and Sorrell, 2007; Ghannoum and Rice, 1999; Lockhart et al., 2017). Importantly, the mechanisms and dynamics of acquired antifungal drug resistance, in vitro or in a patient undergoing antifungal drug therapy, are not fully understood.

The most common human fungal pathogen, Candida albicans, causes nearly 500,000 life-threatening infections each year (Brown and Netea, 2012). Disseminated bloodstream infections of C. albicans have a high mortality rate (15–50%) despite available antifungal therapies (Pfaller et al., 2010; Pfaller et al., 2019). The failure of antifungal drug therapy is likely multifactorial and is compounded by the fungistatic, not fungicidal, mechanisms of most antifungal drugs (Bicanic et al., 2009; Roemer and Krysan, 2014). Additionally, antifungal drug tolerance, the fraction of growth above an individual isolate’s minimum inhibitory concentration (MIC), can cause an inability to effectively clear these fungal infections (Berman and Krysan, 2020). Mechanisms that cause antifungal drug tolerance are not fully understood, but likely include the induction of cell growth and division, core stress response regulators, and cell wall and cell membrane biosynthesis pathways (Berman and Krysan, 2020; Mayer et al., 2013; Onyewu et al., 2004; Rosenberg et al., 2018; Sanglard et al., 2003).

C. albicans and other fungal pathogens exhibit significant karyotype and genome plasticity (Bravo Ruiz et al., 2019; Chibana et al., 2000; Croll and McDonald, 2012; Gerstein et al., 2015; Magee and Magee, 2000; Selmecki et al., 2010; Shin et al., 2007; Sionov et al., 2010; Suzuki et al., 1982; Zolan, 1995). The genome plasticity observed in C. albicans isolates is somatic (asexual), based upon the absence of evidence for a meiotic cell cycle (Alby et al., 2009; Forche et al., 2008; Hull and Johnson, 1999; Magee and Magee, 2000; Tzung et al., 2001), and includes whole genome duplication/reduction, aneuploidy, segmental aneuploidy, and loss of heterozygosity (LOH) (Abbey et al., 2014; Ene et al., 2018; Forche et al., 2008; Forche et al., 2019; Ford et al., 2015; Gerstein et al., 2017; Hickman et al., 2013; Hirakawa et al., 2015; Ropars et al., 2018; Rustchenko-Bulgac, 1991; Selmecki et al., 2006; Todd et al., 2017). From an evolutionary prospective, the genome plasticity of C. albicans (and other fungal pathogens) may dramatically alter the frequency with which beneficial mutations are acquired within a population, resulting in both drug resistance and drug tolerance phenotypes.

Genome plasticity due to amplification or deletion of a chromosome segment, defined herein as copy number variation (CNV), is found across all domains of life (Anderson and Roth, 1977; Beroukhim et al., 2010; Chow et al., 2012; Dulmage et al., 2018; Elde et al., 2012; Riehle et al., 2001; San Millan et al., 2017; Zarrei et al., 2015; Żmieńko et al., 2014). CNVs are highly prevalent in human cancers, resulting in tumorigenesis, metastasis, and increased rates of mortality (Beroukhim et al., 2010; Heitzer et al., 2016; Hieronymus et al., 2018; Shlien and Malkin, 2009; Zack et al., 2013). In Saccharomyces cerevisiae, CNVs of membrane transporters (e.g. HXT6/7, CUP1, GAP1, and SUL1) can provide a strong fitness benefit in nutrient limiting or high-copper environments (Adamo et al., 2012; Brown et al., 1998; Gresham et al., 2008; Gresham et al., 2010; Hull et al., 2017; Lauer et al., 2018; Lin and Li, 2011; Payen et al., 2014; Selmecki et al., 2015). Many CNVs occur due non-allelic homologous recombination (NAHR) between repeat sequences (Chow et al., 2012; Deng et al., 2015; Finn and Li, 2013; Haber and Debatisse, 2006; Hastings et al., 2009; Lobachev et al., 2002; Mizuno et al., 2013; Narayanan et al., 2006; Putnam et al., 2014; Ramocki et al., 2009; Zhao et al., 2014). Some of these CNVs are amplified via a mechanism that relies on short repetitive sequences and aberrant base pairing during replication fork stalling (Brewer et al., 2015; Brewer et al., 2011) or DNA re-replication (Finn and Li, 2013; Green et al., 2010). In both S. cerevisiae and human cancers, extrachromosomal circular DNA (eccDNA) can also yield high copy CNVs (Gresham et al., 2010; Hull et al., 2019; Libuda and Winston, 2006; Møller et al., 2018; Møller et al., 2015; Paulsen et al., 2018; Singh and Wu, 2019). Experimental evolution supports that CNVs can occur and spread rapidly within a population under selection, and competition between distinct CNV lineages (clonal interference) is frequently observed (Lauer et al., 2018; Payen et al., 2014). Additionally, CNVs can increase the rate in which de novo mutations are acquired relative to the rest of the genome, and further alter the mutational and adaptive landscape of viral, bacterial, and eukaryotic organisms (Bayer et al., 2018; Cone et al., 2017; Elde et al., 2012; Otto, 2007; Pavelka et al., 2010; Sun et al., 2009; Yona et al., 2012; Zhou et al., 2011). However, in the absence of selective pressure, CNVs are generally thought to confer a fitness defect to the cell and are removed from the population (Adler et al., 2014; Tang and Amon, 2013). Therefore, the mechanism and dynamics of CNV gain and loss are critical to our understanding of adaptive evolution and pathogenesis.

Antifungal drug stress selects for aneuploidy and CNV formation in diverse human fungal pathogens, including C. albicans, C. auris, and Cryptococcus neoformans (Gerstein et al., 2015; Hwang et al., 2017; Muñoz et al., 2018; Selmecki et al., 2006; Selmecki et al., 2009; Sionov et al., 2010). In C. albicans, a recurrent CNV that amplifies the entire left arm of Chr5 in an isochromosome (i(5L)) is sufficient to cause resistance to azole antifungal drugs (Selmecki et al., 2006; Selmecki et al., 2008). This resistance is due to copy number amplification of two genes on Chr5L: TAC1, a transcriptional activator of the multidrug transporters (Cdr1 and Cdr2), and ERG11, the target of the azole antifungal drugs. i(5L) is frequently identified in clinical isolates and is the only CNV known to cause azole resistance in different genetic backgrounds (Ford et al., 2015; Selmecki et al., 2006; Selmecki et al., 2008; Todd et al., 2019). In addition to copy number amplification, acquisition of non-synonymous, gain-of-function mutations in TAC1 and ERG11 can cause resistance due to constitutive activation of drug efflux pumps and a decreased affinity to the azole drug (Coste et al., 2004; Morio et al., 2010). LOH of these gain-of-function alleles can increase the level of resistance even more dramatically (Coste et al., 2007; Coste et al., 2006; Ford et al., 2015; Sanglard et al., 2003; Selmecki et al., 2008; White, 1997).

Long repeat sequences (65 bp – ~6.5 kb) represent a significant source of genome plasticity in C. albicans isolates obtained both in the presence and absence of antifungal drugs (Todd et al., 2019). All CNV breakpoints and many LOH breakpoints occur at long repeat sequences (Todd et al., 2019). Furthermore, CNV and LOH breakpoints frequently co-occur within the same long repeat sequences. What is not clear is whether a conserved mechanism might link CNV and LOH events during antifungal drug selection, given that both are important mechanisms of acquired antifungal drug resistance.

Through the study of C. albicans isolates subjected to azole antifungal drugs, we have identified a novel mechanism driving the rapid and recurrent formation of high copy CNVs. These CNVs amplify large genomic regions to more than 12 copies per genome and decrease sensitivity to multiple antifungal drugs. CNV formation appears to occur via a dicentric chromosome intermediate and successive breakage-fusion-bridge cycles that are repaired using two distinct long repeat sequences. This mechanism promotes rapid and amplifiable CNV formation during antifungal drug selection. Once the selection is relaxed, cells with the CNV can rapidly return to the progenitor copy number, leaving little evidence that the CNV ever occurred. The transient nature of these CNVs causes phenotypic and population-level heterogeneity that is often observed with clinical isolates in the presence of antifungal drug, including: heteroresistance, trailing growth, and tolerance (Berman and Krysan, 2020; Colombo et al., 2014; Rueda et al., 2017). Ultimately, these CNVs represent a previously uncharacterized, complex mechanism of gene amplification and chromosome plasticity that is exploited during adaptation to antifungal drug stress.

Results

Extensive copy number amplifications occur during adaptation to antifungal stress

To identify mechanisms driving CNV formation during adaptation to antifungal drug stress, we conducted 48 parallel in vitro evolution experiments with four drug-susceptible C. albicans clinical isolates, each representing a distinct genetic background (SC5314, P75016, P75063, and P78042, Supplementary file 1Hirakawa et al., 2015). After 100 generations in physiological concentrations of the most commonly prescribed antifungal drug, fluconazole (FLC, 1 μg/ml) (Felton et al., 2014), the minimum inhibitory concentration (MIC) was determined (See Materials and methods). A sub-set of the FLC-evolved isolates (14/48) that acquired at least a twofold increase in MIC50 relative to their progenitor were selected for whole genome sequencing (WGS). WGS analysis revealed that all (14/14) FLC-evolved isolates acquired one or more whole chromosome and/or segmental chromosome aneuploidies (Figure 1A). Half of the isolates (7/14) acquired novel CNVs (referred to herein as ‘complex CNVs’) that shared several key features: they had high copy numbers (up to 13 copies per genome), occurred entirely within a single chromosome arm (e.g. within Chr1R in AMS4107 and Chr4L in AMS4702), and were not associated with any centromere (CEN) sequence or the C. albicans repetitive element known as the Major Repeat Sequence (MRS, found eight times within the C. albicans genome) (Chibana et al., 1994; Chindamporn et al., 1998; Lephart and Magee, 2006). These complex CNVs amplified chromosome segments that ranged in length from ~164 kb to ~1.02 MB and contained 57–462 ORFs. The remaining isolates acquired either whole chromosome aneuploidies and/or segmental aneuploidies (e.g. i(5L) in AMS4106 and Chr7R in AMS4444) that had typical copy numbers (3–4 copies) and copy number breakpoints (e.g. CEN5 in AMS4106 and MRS7b in AMS4444) that have been observed previously in drug-evolved isolates (Selmecki et al., 2006; Todd et al., 2019), and were not analyzed further.

Figure 1. Complex CNVs are comprised of stair-step amplifications flanked by distinct long inverted repeat sequences.

(A) Whole genome sequence data of FLC-evolved isolates plotted as log2 ratio and converted to chromosome copy number (y-axis, 1–12 copies) and chromosome position (x-axis, Chr1-R) using the Yeast Mapping Analysis Pipeline (YMAP). Complex CNVs amplify >12 copies of a chromosomal region ranging from 164 kb to 1.02 Mb in length. Centromeres indicated with a red arrowhead. Chromosomal positions of all MRS sequences (black dots) and the rDNA array (blue dot) are indicated below AMS4702. The progenitor of each FLC-evolved isolate and the fold increase in FLC MIC50 at 48 hr between the progenitor and the FLC-evolved isolate is indicated. The symmetric stair-step CNV breakpoints for two isolates: (B) Chr1R of AMS4107 and (C) Chr4L of AMS4702. The relative genome sequence read depth is plotted according to chromosome position using R. The left and right side of each CNV (indicated with blue and orange lines along the x-axis of each chromosome) are expanded for higher resolution (left and right lower panels). In both isolates (AMS4107 and AMS4702), the CNV is flanked by two, distinct long inverted repeat sequences (blue and orange arrows) that do not share homology. The highest copy number amplification occurs between the two, distinct inverted repeats. All copy number breakpoints and long inverted repeat sequence details are found in Supplementary file 2. All genes found within the amplified regions are found in Supplementary file 4. Repeat numbers refer to Supplementary file 2 from Todd et al., 2019.

Figure 1.

Figure 1—figure supplement 1. Recurrent CNV breakpoints located on Chr3R amplify MRR1.

Figure 1—figure supplement 1.

The relative genome sequence read depth is plotted according to chromosome position using R. The left and right side of the highest copy number region of the CNV (indicated with blue and orange lines along the x-axis of each chromosome) are expanded for higher resolution (left and right lower panels). Isolates (A) AMS4104, (B) AMS4397, and (C) AMS4105 all amplify a ~208 kb region on Chr3R that includes MRR1 (asterisks), the transcriptional regulator of the multidrug transporter Mdr1.
Figure 1—figure supplement 2. Complex CNVs increase chromosome size.

Figure 1—figure supplement 2.

Contour-clamped homogenous electric field (CHEF) electrophoresis of each progenitor (in bold) and FLC-evolved isolates. Chromosomes (Chr) are labeled to the left of the gel (approximate Chr sizes based on SC5314); some progenitors have separable homologs of Chrs 4–7. An increase in Chr4 size due to the Chr4L CNV (Figure 1) is indicated for two FLC-evolved isolates (AMS4702 and AMS4444, asterisks) relative to their progenitor (SC5314 and P78042, respectively). The Chr4L CNV increases the size of Chr4 from ~1.6 Mb to ~3.4 Mb for AMS4702 and to ~2.8 Mb for AMS4444, and results in the Chr4 band migrating with Chr1 and ChrR, respectively. Approximate chromosome size increases due to each CNV are in Supplementary file 6.
Figure 1—figure supplement 3. Complex CNVs are predominantly found in homozygous sequence.

Figure 1—figure supplement 3.

Whole genome sequence data of progenitor isolates in bold (SC5314, P75016, P75063, and P78042) and independent, FLC-evolved isolates (AMS4702, AMS4397, AMS4104, AMS4105, AMS4016, AMS4017, and AMS4444) plotted as log2 ratio and converted to chromosome copy number (y-axis) and chromosome position (x-axis) using YMAP. Heterozygous loci indicated with gray hashing. Over half (7/10) of the CNVs detected occur in regions that are already homozygous in the progenitor. For example, allele ratios indicate that the CNV breakpoints on Chr1 (AMS4105 and AMS4107), Chr3 (AMS4397, AMS4104, and AMS4105), Chr5 (AMS4106), and Chr7 (AMS4444) occurred at regions that were homozygous or were heterozygous-to-homozygous breakpoints in the progenitor. (1/10) of the CNVs coincide with novel loss of heterozygosity (LOH) breakpoints (Chr4, AMS4702). (2/10) CNVs occurred in heterozygous sequence, and allele ratios scale with copy number (Chr 4, AMS4106 and AMS4444). These data suggest that CNVs, when heterozygous, are from a single haplotype, and that many of these CNVs may be transient events that previously drove LOH in the progenitor.

Complex CNVs are flanked by distinct long inverted repeat sequences

To identify the mechanism driving the formation of the novel complex CNVs, we determined the copy number breakpoints associated with each of the complex CNVs using a combination of read depth and allele ratio analyses. All copy number breakpoints occurred within 2 kb of one of the 1974 long repeat sequences identified previously by Todd et al., 2019 (Supplementary file 2). However, unlike previous observations, these complex CNVs were comprised of a two-sided, stair-step amplification pattern that involved at least two distinct long inverted repeat sequences (Supplementary file 2). Accordingly, each long repeat sequence was associated with two distinct copy number changes, generating regions with different degrees of amplification, with the highest copy number always flanked by lower copy numbers (Figure 1B and C, Supplementary file 6). For example, a complex CNV on Chr1R (in AMS4107) comprised a ~218 kb region amplified to nine copies, flanked on the left and right by a region of variable length (the intra-repeat spacer length) that was amplified to six copies, that, in turn, was surrounded on both sides by a region of 2N copy number (the basal chromosome copy number). Surprisingly, the same symmetric stair-step copy number pattern (2-6-9-6-2 copies) was observed on two different chromosomes (Chr1 and Chr4) in two different genetic backgrounds (AMS4107 and AMS4702), indicating that the mechanism is neither chromosome-specific nor strain-specific (Figure 1B and C). Complex CNVs with asymmetric stair-step copy numbers also occurred due to an additional breakpoint in a third distinct long inverted repeat sequence (e.g. 2-4-6-5-3-3-2 copies in AMS4105 and 2-3-3-6-9-6-2 copies in AMS4397), that also increased the length of the CNV (see Materials and methods for determination of copy number intervals, Supplementary file 6). Of the eight complex CNVs, the highest copy number identified (2-7-13-7-2 copies) occurred in AMS4104 on Chr3R. These findings indicate that the mechanism driving the formation of these complex CNVs uses long repeat sequences, that the amplified number of copies is variable, and that both odd and even numbers of amplified copies can be detected.

Amplification of the same chromosomal region also occurred in isolates with different genetic backgrounds and from independent evolution experiments (Figure 1A). These recurrent CNVs highlight genes that are likely under selection during adaptation to FLC. For example, recurrent CNV breakpoints on Chr1R occurred at Repeats 57 and 58 in AMS4107 and AMS4105, and recurrent breakpoints on Chr3R occurred at Repeats 134 and 136 in AMS4105, AMS4397, and AMS4104 (Figure 1—figure supplement 1, Supplementary file 2). Interestingly, while the CNV breakpoints on Chr4L were not recurrent and instead occurred at different long repeat sequences, all three of the CNVs amplified a common ~118 kb region in AMS4106, AMS4444, and AMS4702 (Supplementary file 2). Therefore, during FLC selection, complex CNV formation results in copy number amplification of recurrent chromosomal regions.

All the long inverted repeat sequences associated with the complex CNVs were contained within a single chromosome arm (intra-chromosome arm). The occurrence of intra-chromosome arm inverted repeat sequences is relatively low (77/233) within the C. albicans genome (excluding MRSs and ORFs which contain complex embedded tandem repeats, see Todd et al., 2019). The repeat sequences found at these complex CNV breakpoints were among the longest (median 1513 bp) repeats, and had among the highest median shared sequence identity (96.1%) of all long repeat sequences found throughout the genome (median 516 bp, 95.1% median shared sequence identity), which is similar to repeats associated with breakpoints resulting in CNV, LOH and large chromosomal inversions in C. albicans isolates obtained in the presence and absence of antifungal drugs (Todd et al., 2019).

To ask if these complex CNVs were intra-chromosomal or extra-chromosomal amplifications (e.g. eccDNA that appear in in budding yeast and cancer cells; Hull et al., 2019; Møller et al., 2018; Møller et al., 2015; Paulsen et al., 2018; Wu et al., 2019), we used CHEF karyotype analyses. Separation of Chrs 4–7 identified a dramatic increase in Chr4 size in two isolates (AMS4702 and AMS4444) with Chr4L CNVs relative to their progenitors and indicates that these CNVs are intra-chromosomal, rather than extra-chromosomal, amplifications (Figure 1—figure supplement 2). Because detection of CNVs on Chr1 and Chr3 was obviated by the large size of these chromosomes, restriction digest was used to characterize these intra-chromosomal amplification events (see below, e.g. Figure 3). Therefore, while we cannot completely rule out the possibility of eccDNA amplifications in some of the isolates, the increased chromosome sizes (Figure 1—figure supplement 2) support the idea that the complex CNVs are intra-chromosomal amplification events.

Surprisingly, these CNVs tended to occur within chromosomal regions that were homozygous in the progenitor isolates, making it impossible to determine which haplotype was amplified in the CNVs (Figure 1—figure supplement 3). Nonetheless, for the CNVs from heterozygous progenitor sequences, the amplifications derived from only one haplotype and the allele ratio scaled with copy number (e.g. the percent majority allele was ~50–62–71–64–50 for a 2-3-4-3-2 copy CNV in AMS4106). This supports that the complex CNVs arose via a stair-step mechanism that amplified just one haplotype.

Complex CNVs increase multidrug fitness and tolerance

While bacteria usually exhibit a fitness tradeoff between drug resistance and fitness (Bagel et al., 1999; Basra et al., 2018; Melnyk et al., 2015), the situation is far less clear in fungal pathogens. To explore this issue comprehensively, we performed growth curve and MIC analyses across the genetically diverse isolates that had been evolved in vitro. In rich medium, the growth rate and maximum OD600 was similar between all FLC-evolved isolates and their progenitors (Figure 2A–D, left panels), with several notable increases in lag phase length (growth curve summary statistics provided in Supplementary file 3). In the presence of FLC (1 μg/ml), the growth rate and maximum OD600 were increased for all FLC-evolved isolates relative to their progenitors, and one isolate (AMS4444) grew better in FLC than in rich medium (Figure 2A–D, right panels, Supplementary file 3). While each progenitor and FLC-evolved isolate had unique growth trajectories, these observations support, in general, that the complex CNVs provided increased fitness in the presence of drug without a major cost to fitness in the absence of drug.

Figure 2. Complex CNVs increase multidrug fitness and tolerance.

(A–D) 36 hr growth curve analysis in the absence (YPAD, left) and presence of FLC (YPAD + 1 μg/ml FLC) for each progenitor (black) and FLC-evolved isolate with a complex CNV. Average slope and standard error of the mean for three biological replicates is indicated. (E–F) Heat map of isolate growth (OD600 at 24 and 48 hr) in two-fold increasing concentrations of the azole antifungal drugs (E) fluconazole (FLC) and (F) miconazole. The drug concentration at which 50% of growth is inhibited (MIC50) is denoted with a yellow line on the heat map. Each heat map represents the average of three independent MIC50 assays. Supra-MIC growth (SMG), a measurement of tolerance, was calculated as the average growth at 48 hr above the MIC50 at 24 hr divided by the growth at 48 hr in no drug (see Materials and methods, Figure 2—source data 1).

Figure 2—source data 1. Minimum inhibitory concentration raw data.

Figure 2.

Figure 2—figure supplement 1. Complex CNVs increase tolerance and reduce susceptibility to multiple azole drugs.

Figure 2—figure supplement 1.

Heat map of OD600 values at 24 and 48 hr in two-fold increasing concentrations of the following azole antifungal drugs: (A) fluconazole (FLC); (B) miconazole; (C) itraconazole; (D) ketoconazole; and (E) posaconazole. The MIC50 is denoted with a yellow line. Each heat map represents the average of three independent MIC50 assays. Independently evolved isolates are grouped by progenitor (bold). Supra-MIC growth (SMG), a measurement of tolerance, was calculated as the average growth at 48 hr for all wells above the MIC50 at 24 hr divided by the growth at 48 hr in no drug (see Materials and methods, Figure 2—source data 1).

The sub-inhibitory concentrations of FLC used in our evolution experiment and the fungistatic nature of azole antifungal drugs can select for mutants that exhibit drug tolerance: the ability to grow at drug concentrations above the MIC50 after 24 hr (Berman and Krysan, 2020; Delarze and Sanglard, 2015; Rosenberg et al., 2018; Sanglard et al., 2003). We measured tolerance to FLC and four other azole drugs (miconazole, itraconazole, ketoconazole, and posaconazole) as supra-MIC growth (SMG), which is calculated from the average growth (OD600) at 48 hr for all wells above the MIC50 at 24 hr (Berman and Krysan, 2020; Rosenberg et al., 2018). Four of the seven isolates with complex CNVs had higher SMG levels than their progenitors in FLC, and all seven had higher SMG levels in miconazole (Figure 2E and F, Figure 2—figure supplement 1). AMS4444 and AMS4105 had the highest SMG level in all azoles tested (0.53–0.73) and exhibited the highest growth rates and maximum OD600 in 1 μg/ml FLC (Figure 2C–D). Therefore, all isolates that acquired a complex CNV had increased fitness (growth rate and MIC50) and increased tolerance (SMG) to one or more azole antifungal drugs, and several isolates had increased tolerance to all azoles tested.

Genes involved in drug resistance and tolerance are amplified in recurrent CNVs

To identify genes located within the CNVs that could be driving the increased fitness in azole drugs, we first characterized genes known to cause drug resistance. We started by looking at genes with the potential to encode efflux pumps. We found that MRR1, the transcriptional regulator of the multidrug efflux pump Mdr1, was amplified by up to 13 copies in the three isolates with a Chr3R CNV (Figure 1—figure supplement 1). Gain-of-function mutations in MRR1 can cause constitutive upregulation and multidrug resistance (Dunkel et al., 2008; Morschhäuser et al., 2007; Schubert et al., 2008); copy number amplification alone was not previously shown to cause resistance. Additionally, multidrug transporters CDR1 and CDR2 map to Chr3R near MRR1, and these genes were amplified within the same CNV as MRR1 in two isolates (AMS4105 and AMS4397). MRR1 was always amplified at the highest copy number of the CNVs, while CDR1 and CDR2 were amplified to lower copy numbers. Finally, two other genes within the complex CNVs, QDR2 (Chr3R) and orf19.4889 (Chr1L), both encode predicted major facilitator superfamily (MFS) membrane transporters that may have a role in antifungal drug efflux.

Next, we searched for genes that are required for membrane and cell wall integrity, calcium and iron availability, and core stress response pathways which are likely to be important for antifungal tolerance (Berman and Krysan, 2020; Garnaud et al., 2018; O'Meara et al., 2017; Onyewu et al., 2004; Rosenberg et al., 2018; Sanglard et al., 2003; Taff et al., 2013). The calcineurin-regulated transcription factor CRZ1, involved in maintenance of membrane integrity and antifungal drug tolerance (Onyewu et al., 2004; Sanglard et al., 2003), also maps to Chr3R and was amplified along with MRR1, CDR1, and CDR2 in the same two isolates mentioned above, both of which had increased tolerance and resistance to multiple azole drugs (AMS4105 and AMS4397, Supplementary file 4). Other genes that encode other stress response proteins (HSP70, CGR1, ERO1, TPK1, ASR1, and PBS2) and proteins involved in membrane and cell wall integrity (CDR3, NCP1, ECM21, MNN23, RHB1 and KRE6) were amplified in the CNVs (Supplementary file 4).

No correlation between the copy number of specific genes and the MIC50 or SMG observed was evident. However, since these CNVs occurred in different genetic backgrounds, that differ by ~100,000 unique SNVs, the amplification of certain alleles is likely to affect fitness differently in each isolate. For isolates from the same genetic background, the copy number of each CNV and presence of additional chromosome aneuploidies may further impact fitness. The most striking increase in FLC tolerance (SMG 0.54) was observed for an isolate (AMS4105) that acquired two different CNVs: amplification of Chr3R (containing MRR1, CDR1, CDR2, QDR2, CRZ1, etc.) and Chr1L (containing ORF19.4889, etc). A different isolate (AMS4104) from the same genetic background had the same MIC50 at 24 hr (as AMS4105), but no increase in tolerance (SMG = 0.13) relative to the progenitor (SMG = 0.16). AMS4104 acquired only a Chr3R CNV containing MRR1, but no amplification of CDR1, CDR2, QDR2, CRZ1, or the Chr1L CNV. Therefore, amplification of the MRR1-containing CNV correlates with an increase in the MIC50, while amplification of additional genes in AMS4105 on Chr3R and Chr1L (including CDR1, CDR2, and CRZ1) appear to have a greater impact on tolerance.

To ask if any gene functions were enriched within the complex CNVs, we performed gene ontology (GO) analysis (Supplementary file 4). The cellular component ‘nuclear microtubule’ was the only GO term significantly enriched for all genes located within the complex CNVs (p<0.008, Bonferroni correction for multiple comparisons). The ‘nuclear microtubule’ term included genes that encode microtubule-binding proteins involved in spindle elongation, organization, and stabilization (ASE1, KIP3, and BIM1), chromatin remodeling (ISW2), and ribosome biogenesis (NEP1) (Côte et al., 2009; Enjalbert et al., 2006; Eschrich et al., 2002; McCoy et al., 2015; Nobile et al., 2003; Nobile et al., 2012; Singh et al., 2011; Tuch et al., 2008Supplementary file 4). In summary, the complex CNVs amplified genes known to have a direct role in drug resistance (MRR1, CDR1, and CDR2) and drug tolerance (CZR1), as well as other roles that may be under selection during adaptation to antifungal drug stress, including maintenance of mitotic spindle function.

Expandable CNVs generated through a step-wise amplification of a dicentric chromosome

The identification of recurrent CNV breakpoints in different genetic backgrounds raised the possibility that a common mechanism was driving the formation of these complex CNVs. To further address the mechanism of formation and to understand the impact of copy number on fitness, we used a set of isogenic isolates obtained from an agar plate instead of liquid cultures. These four single colony isolates (AMS3051, AMS3052, AMS3053, and AMS3054) were obtained from the same progenitor (AMS3050) after 120 hr growth on a single miconazole (20 μg/ml) agar plate (Mount et al., 2018). Prior whole genome sequencing analysis of these colonies identified a shared CNV of Chr3L (Mount et al., 2018). To test the hypothesis that these single colonies represented different outcomes from the same recombination event, due to the short exposure to miconazole and similarity in karyotypes, we first performed read depth analysis to characterize the CNV breakpoints. All four colonies were monosomic from the Chr3L left telomere to a shared CNV breakpoint at a long inverted repeat (Repeat 124 [blue lines], Figure 3A). Strikingly, the major difference between the four colonies was the maximum number of copies (3–14 copies) of the adjacent ~146 kb region on Chr3L. This region (as in Figure 1) was flanked by two distinct long inverted repeat sequences resulting in a stair-step amplification in AMS3051 and AMS3054 (Repeats 124 [blue lines] and Repeats 127 [orange lines], Figure 3A, Figure 3—figure supplement 1). In one isolate (AMS3052), the region of variable amplification extended beyond Repeat 127 to the centromere of Chr3 (CEN3). Importantly, the MIC50 of the miconazole-evolved colonies increased as the maximum copy number of this complex CNV increased (Figure 3B), supporting that in an isogenic background the increase in copy number directly correlates with an increase in MIC.

Figure 3. Complex CNVs are rapidly expandable in the presence of antifungal drug.

(A) Whole genome sequence data of progenitor isolate (AMS3050) and miconazole-evolved single colonies (AMS3053, AMS3054, AMS3052, and AMS3051) plotted as in Figure 1A. All four colonies are monosomic from the Chr3L telomere to a CNV breakpoint at Repeat 124 (blue lines). Stair-step complex CNVs (3 to 14 copies per genome) occur on Chr3L between two distinct long inverted repeat sequences: Repeat 124 (blue lines) and Repeat 127 (orange lines) (detailed in Figure 3—figure supplement 1). All four colonies are trisomic for the Chr3 centromere (CEN3, black circle) and all of Chr3R. (B) The MIC50 increases with the copy number of the complex CNV. Heat map of isolate growth (OD600 at 48 hr) in twofold increasing concentrations of miconazole. The MIC50 is denoted with a yellow line. Each heat map represents the average of three independent MIC50 assays (Figure 2—source data 1). (C) Schematic of the homologous chromosomes in AMS3051. The full length (gray) and dicentric CNV-containing (black) homologs with the positions indicated for CEN3 (circle), the long repeat sequences (blue and orange lines), and SacII cut sites (dashed lines). Four regions (I–IV) that support a breakage-fusion-bridge mechanism for the formation of complex CNVs (see main text for details). (D) Allele ratio plot of all heterozygous loci located within and flanking the complex CNV for AMS3050 and AMS3051. Allele ratio plots for all isolates are in Figure 3—figure supplement 2. (E) SacII-digested CHEF karyotype of the progenitor and miconazole-evolved isolates (first lane of undigested AMS3050 is shown for relative size). The SacII digest isolates the region with variable copy number and CEN3 (schematic at far right). CHEF gel stained with ethidium bromide (left panel) and analyzed by Southern blot using a DIG-labeled probe to orf19.344, located within the complex CNV (middle panel), and CEN3 (right panel). Both Southern blot probes detect a novel band that increases in size as the complex CNV increases in copy number.

Figure 3.

Figure 3—figure supplement 1. The Chr3L CNV is flanked by two, distinct inverted repeat sequences.

Figure 3—figure supplement 1.

Relative genome sequence read depth plotted according to chromosome position for (A) the progenitor AMS3050 and (B) the miconazole-evolved isolate AMS3051. The complex CNV in AMS3051 is flanked by two, distinct long inverted repeat sequences: Repeat 124 (blue arrows/data points) and Repeat 127 (orange arrows/data points). The highest copy number amplification occurs between the two non-homologous inverted repeats. All features of the CNV breakpoints are detailed in Supplementary file 2; repeat numbers refer to Supplementary file 2 from Todd et al., 2019.
Figure 3—figure supplement 2. CNVs are intra-chromosomal amplifications between two distinct inverted repeat sequences.

Figure 3—figure supplement 2.

Allele ratio plot of all heterozygous loci located within and flanking the CNV breakpoints (blue and orange arrows indicate the inverted repeat sequences at the CNV breakpoints) and the CEN3 (black dot). Allele ratios shift from +0.52/- 0.48 in the progenitor AMS3050 to +0.66/- 0.34 in AMS3053, +0.77/- 0.23 in AMS3054, +0.79/- 0.21 in AMS3052, and 0.90+/- 0.10 in the 14 copy CNV region of the evolved isolate AMS3051. The specific C. albicans chromosome A/B haplotypes were reported previously (Mount et al., 2018).

The isogenic isolates contained five genomic features that provide clues concerning the mechanism of complex CNV formation on Chr3L (Figure 3C): I) Monosomy (and thus LOH) extending from Repeat 124 to the telomere; II) Complex CNVs with a maximum copy number that varies between the individual colonies and amplifies unique sequences between two distinct long inverted repeats (Repeats 124 and 127); III) Trisomy of sequences extending from Repeat 127 to CEN3; IV) Trisomy of CEN3 and all of Chr3R; and V) Amplification of a single haplotype in the complex CNV of Chr3L and throughout the trisomic region of Chr3R (Figure 3D, Figure 3—figure supplement 2) (although this is not detectable in the region distal to Chr3R position 896,538, because this region is already homozygous in the progenitor). From these observations, we hypothesized that one homolog of Chr3 was intact and includes the monosomic portion of Chr3L, while the other homolog has formed a dicentric molecule that promotes breakage-fusion-bridge (BFB) cycles that result in the complex, stair-step amplifications between and within the long inverted repeats (Figure 3C).

If this hypothesis is correct, we should be able to detect dicentric chromosome intermediates by CHEF gel karyotype analysis. Because an intact dicentric Chr3 would be too large to resolve, we analyzed SacII-digested chromosomal DNA. SacII sites fall to the left of the complex CNV of Chr3L (within the monosomy) and to the right of CEN3 yielding a fragment that should be ~480 kb in the AMS3050 progenitor (Figure 3C and E). Indeed, the progenitor and all four evolved isolates had a band of ~480 kb (Figure 3E). In addition, the four evolved isolates had an additional band of increased size (~854 kb to ~2.16 Mb) consistent with the level of amplification in these isolates. This high-molecular-weight band hybridized to two different probes: an ORF found within the most amplified region of the Chr3L CNV (orf19.344) and CEN3. Thus, the size and content of the SacII-digested region including CEN3 is consistent with the idea that the fragment includes a dicentric chromosome that is linked via the region containing the different-sized complex CNVs found among the AMS3050 derivatives (Figure 3C and E).

Recombination occurs between long inverted repeats leading to CNV formation

Models of BFB in both fungal and human cells support that dicentric chromosomes form via non-allelic homologous recombination (NAHR) between inverted repeat sequences (Croll et al., 2013; Hermetz et al., 2014; Notta et al., 2016). We used Oxford Nanopore Technology long-read sequencing to test the hypothesis that NAHR between repeat sequences was involved in the formation of the complex CNV on Chr3L in AMS3051 (Figure 4A–C) . Structural variants, indicators of recombination products not identified in the reference genome, were identified in AMS3051 using split-read alignments, read mismatching, and read depth analyses (see Materials and methods). One structural variant was detected at Repeat 124 (e.g. Figure 4B), and the other structural variant was detected at Repeat 127 (e.g. Figure 4C). Each structural variant combined unique (non-repeat) sequences, that were located up to ~100 kb apart in the reference genome, into a single long-read of only 8 kb – 10 kb in length. Approximately 50 unique long-reads supported these structural variants and each long-read included a single copy of either Repeat 124 or Repeat 127. The individual long-reads switched between the complement and reverse complement orientation (relative to the reference) within the long repeat sequence (Repeats 124 and 127, Figure 4B and C). These long-reads represent fold-back inversions that were presumably mediated by NAHR between distant copies of the long repeat sequences. These observations, together with the intra-chromosomal CNV expansions detected by CHEF (Figure 3E), suggest that the complex CNV located on Chr3L occurs via an accordion-like expansion.

Figure 4. Long-read sequencing reveals recombination products involved in the formation of complex CNVs.

Figure 4.

(A) Relative Illumina read depth for the miconazole-evolved isolate AMS3051 plotted by chromosome position (data as in Figure 3—figure supplement 1) indicating the presence of two inverted repeats (Repeat 124 (blue line) and Repeat 127 (orange line)) flanking a complex CNV on Chr3L. Repeat numbers refer to Supplementary file 2 from Todd et al., 2019. In the reference genome, Repeat 124 (B), consists of two inverted copies (~99% sequence identity) that are ~3 kb in length that are ~11 kb apart and Repeat 127 (C) consists of two inverted copies (~99% sequence identity) that are ~4 kb in length and located ~100 kb apart. Analysis of long-read sequences from AMS3051 identified structural variants relative to the reference genome at both Repeats 124 and 127. A single representative long-read (bottom purple line) aligned to the reference genome (top black line) is shown for Repeat 124 (B) and Repeat 127 (C). The long-read contains unique sequences that are separated by up to ~100 kb in the reference genome. Green-colored areas indicate alignment to the complement strand and gray-colored areas indicate alignment to the reverse complement strand of the reference genome. The transition between complement/reverse complement occurs within the repeat sequence. Schematics of both Repeat 124 and Repeat 127 indicate the formation of a fold-back inversion and non-allelic homologous recombination (NAHR, black and purple dashed lines) between repeat copies on sister chromatids, which could generate a dicentric chromosome. Alignment of the recombination product is inferred to produce the long-read that was detected (purple bar within the dicentric chromosomes). Each structural variant was supported by ~50 long-read sequences (see Materials and methods). All features of the CNV breakpoints are detailed in Supplementary file 2.

Loss of the CNVs and subsequent LOH in the absence of antifungal drug selection

Highly amplified CNVs are expected to be subject to recombination events that reduce copy number, especially if selection for the extra gene copies is relaxed. To determine the stability of the Chr3L CNV in the absence of antifungal drug, we isolated single colonies on rich medium after 72 hr. Heterogeneous populations of large and small colonies were observed for each of the four miconazole-evolved isolates (Figure 5A and B). Both a large and small colony isolated from AMS3051 were plated for single colonies on rich medium: the large colony gave rise to similarly large colonies (AMS3092), while the small colony continued to give rise to a heterogeneous population of large (AMS3093) and small (AMS3094) colonies (Figure 5A and B). WGS and read depth analysis supported that the small colony phenotype in the absence of miconazole was due to a fitness defect associated with the dicentric chromosome and/or the monosomic portion of Chr3L. In contrast, both large colonies (AMS3092 and AMS3093) had resolved the dicentric chromosome and regained the disomic portion of Chr3L (Figure 5A, Figure 5—figure supplement 1). In one case (AMS3092), the dicentric chromosome underwent a recombination event that maintained the complex CNV and heterozygosity across CEN3 and Chr3R. In the other case (AMS3093), the dicentric chromosome underwent a recombination event at CEN3 that returned this isolate to a euploid genotype (Figure 5—figure supplement 1) and homozygosed all of Chr3L (Figure 5—figure supplement 2). In both examples, the dicentric chromosome was never completely lost, but had recombined to resolve the dicentric.

Figure 5. Complex CNVs resolve in the absence of antifungal drug by eliminating the dicentric chromosome.

(A) Representative images of the progenitor (AMS3050) and the four miconazole-evolved isolates (AMS3053, AMS3054, AMS3052, and AMS3051) containing the Chr3L CNV grown on YPAD. Copy number of Chr3 (from Figure 3) shown to the right of the plate images. The notch in Chr3 is CEN3. Representative images below AMS3051 of single colonies derived from either a small (blue) or large (black) colony of AMS3051 on rich medium: The small colony gave rise to AMS3094 and AMS3093 and the large colony gave rise to AMS3092. Copy number of Chr3 shown below the plate images. Whole genome sequencing data are provided in Figure 5—figure supplement 1. (B) Colony size analysis using ImageJ (see Materials and methods), n > 113 (up to n = 300) single colonies, three biological replicates. (C) Heat map of OD600 values taken at 48 hr in twofold increasing concentrations of miconazole. The MIC50 is denoted with a yellow line. Each heat map represents the average of three independent MIC50 assays (Figure 2—source data 1). (D) CHEF of the parental strain and miconazole-evolved isolates. Whole genomic DNA was digested with SacII to isolate the region containing the CNV and CEN3 (schematic to right). CHEF gel stained with ethidium bromide (left panel) and analyzed by Southern blot with DIG-labeled probes to orf19.344 within the CNV (middle panel) and CEN3 (right panel).

Figure 5.

Figure 5—figure supplement 1. Loss of Chr3 CNV correlates with reduction of MIC50 to miconazole.

Figure 5—figure supplement 1.

Whole genome sequence data of miconazole-evolved isolates plotted as log2 ratio and converted to chromosome copy number (y-axis) and chromosome position (x-axis) using the Yeast Mapping Analysis Pipeline (YMAP). (A) The progenitor strain (AMS3050) and the four evolved isolates (AMS3053, AMS3054, AMS3052, and AMS3051) with the Chr3 CNV and Chr7 trisomy. MIC50 at 48 hr to miconazole shown to the right. (B) The three single colonies (AMS3094, AMS3092, and AMS3093) derived from AMS3051. MIC50 shown to the right.
Figure 5—figure supplement 2. Additional loss of heterozygosity can occur during resolution of complex CNVs.

Figure 5—figure supplement 2.

Allele ratio plots of all heterozygous loci on Chr3 for (A) the CNV-containing progenitor isolate AMS3051, and three single colonies isolated from AMS3051 in the absence of antifungal drug (B–D). Location of the long inverted repeats (blue and orange arrows) and CEN3 (black dot) are shown above each plot. The progenitor to AMS3051 is homozygous on Chr3R from position 896,538 to the Chr3R telomere. (B) AMS3094 maintains the entire Chr3 CNV structure and allele frequency ratios similar to AMS3051. (C) AMS3092, which kept the complex stair-step CNV on Chr3L but became disomic for both Chr3R and the telomere proximal region of Chr3L, returned to a ~50% allele frequency ratio between the CNV and CEN3 while remaining homozygous for the telomere proximal disomic region of Chr3L. (D) Recombination near CEN3 resulted in LOH and loss of the CNV. The euploid strain AMS3093 became homozygous for the entire left arm of Chr3 while remaining heterozygous to the right of CEN3.
Figure 5—figure supplement 3. Dicentric Chr3 is stable in the presence of antifungal drug.

Figure 5—figure supplement 3.

(A) Representative image of AMS3053 single colonies on a plate containing 20 μg/ml miconazole for 72 hr. (B) Colony size heterogeneity of AMS3053 in the absence and presence of 20 μg/ml miconazole. Colony size determined using ImageJ, n = 113–115 single colonies obtained from three individual agar plates (see Materials and methods). (C) CHEF of the progenitor (AMS3050), the initial miconazole-evolved isolate with a dicentric Chr3 (AMS3053), and eight randomly selected single colonies (AMS3053_A-H) derived from AMS3053 after 72 hr on 20 μg/ml miconazole (plate pictured in A). Karyotype analysis performed as in Figure 5D: whole genomic DNA was digested with SacII to isolate the region containing the dicentric Chr3 band (schematic to right). CHEF gel stained with ethidium bromide (left panel) and analyzed by Southern blot with DIG-labeled probes to orf19.344 within the CNV (right panel).

The impact of CNV loss on fitness was determined for the three single colonies derived from AMS3051 (Figure 5C). The highest MIC50 (4 μg/ml miconazole) was observed for both isolates with the CNV on a dicentric chromosome (AMS3051 and AMS3094). Surprisingly, the MIC50 decreased (2 μg/ml miconazole) for the isolate that retained the complex CNV (AMS3092) but had become euploid for all other regions on Chr3. One possibility is that, loss of the extra copy of Chr3R (which contains the multidrug resistance and tolerance genes described above: MRR1, CDR1, CDR2, and CRZ1) reduced the MIC50 of this strain. Therefore, the decrease in MIC50 (between AMS3051 and AMS3092) may be due to reduced copy number of these genes.

As expected, the MIC50 was lowest (1 μg/ml miconazole) for the isolate that returned to the euploid genotype (AMS3093). This MIC50 was still above the MIC50 of the euploid progenitor AMS3050 (0.5 μg/ml miconazole) despite the lack of de novo SNVs between the two isolates (see Materials and methods). The difference in MIC50 (between AMS3050 and AMS3093) may be due to the additional LOH of Chr3L that occurred during the resolution of the dicentric chromosome in AMS3093 (Figure 5D, Figure 5—figure supplement 2).

Finally, under constant antifungal drug selection (20 μg/ml miconazole), the dicentric chromosome appeared to be highly stable by both colony morphology and karyotype analysis (Figure 5—figure supplement 3). Thus, under continued antifungal drug selection the dicentric chromosome is maintained at the population and single-cell level. In contrast, removal from the selection pressure promotes the loss/resolution of the dicentric chromosome.

Discussion

This study identifies a novel mechanism for generating beneficial CNVs during adaptation to physiologically relevant concentrations of azole antifungal drugs. The formation of these complex CNVs is rapid, expandable, and reversible. Recurrent CNV breakpoints occur in clinical isolates with diverse genetic backgrounds that are exposed to azole antifungal drugs and can increase fitness across multiple azole drugs. The heterozygous diploid genome of C. albicans has enabled us to determine the mechanism of CNV formation from population-level and single-cell karyotype analyses. Ultimately, the expansion and contraction of these CNVs affects the rate and dynamics in which antifungal drug resistance and tolerance is acquired, and provides a plausible explanation, independent of whole chromosome aneuploidy, for the phenotypic heterogeneity frequently reported during antifungal drug susceptibility testing (e.g., tolerance or trailing growth and heteroresistance) (Ben-Ami et al., 2016; Qiao et al., 2017; Sionov et al., 2009).

Model for complex CNV formation

The complex CNVs amplified recurrent chromosome regions and were associated with increased fitness in azole antifungals. Short-read and long-read whole genome sequencing revealed that all of the complex CNVs have copy number breakpoints that occur within distinct long inverted repeat sequences; in other words, the amplified regions reside between sets of different inverted repeat sequences. Based on these breakpoint sequences, we propose a model of NAHR and breakage-fusion-bridge (BFB) cycles for the formation and resolution of complex CNVs in C. albicans (Figure 6). For example, to generate the Chr3L complex CNV, a DNA double-strand break (DSB) occurs (the exact location of the DSB is not known) between the telomere and the telomere-proximal inverted repeat (Figure 6A and B, Repeat 124, blue arrows). This DSB generates a telomere proximal acentric fragment that is lost during future cell divisions. After DNA replication, NAHR between the long inverted repeat sequences (Figure 6A and B, Repeat 124, blue arrows) located on sister chromatids generates a dicentric chromosome and amplification of all sequence from the site of NAHR to the telomere on Chr3R (Croll et al., 2013; Lang et al., 2013; Ramakrishnan et al., 2018; Stimpson et al., 2012). Alternatively, the dicentric chromosome could be formed via an intra-chromosomal fold-back-like mechanism between repeat copies that primes break induced replication (BIR) (Narayanan et al., 2006; Rattray et al., 2005). During cytokinesis, the dicentric chromosome bridge is broken preferentially at a centromere-proximal location due to closure of the actomyosin ring (Lopez et al., 2015); the longer, monocentric chromosome fragment is followed in the model. During the next cell cycle, DNA replication generates two sister chromatids that again undergo NAHR, this time between the two copies of the centromere-proximal long inverted repeat sequence (Figure 6A and B, Repeat 127, orange arrows), which generates a second dicentric chromosome and an amplification of all sequence from the site of the second NAHR event to the telomere on Chr3R. These dicentric BFB cycles can result in heterogeneous outcomes that are beneficial in the presence or absence of antifungal drug selection including: further amplification of the repeat and unique sequences (Figure 6C); NAHR with a repeat on the other homolog, resulting in resolution of the dicentric and isolation of the CNV on a monocentric chromosome (Figure 6D, top); and recombination (likely BIR) that occurs with the other homolog at a position centromere proximal to the CNV and continues to the Chr3L telomere, resulting in loss of the complex CNV and homozygosis of Chr3L (Figure 6D, bottom). Example symmetric stair-step copy number amplifications generated by these recombination events and the presence of fold-back inversions that arise during NAHR between repeat sequences are proposed in the model (Figure 6C and D). We propose a similar model for asymmetric stair-step copy number amplifications that can form due to recombination events involving three (instead of two) distinct inverted repeat sequences (Figure 6—figure supplement 1). Finally, the Chr3L CNV expansions (Figure 3, during antifungal selection) and contractions (Figure 5, when selection is relaxed) were detected after formation of a single colony and underscore the dynamic potential of these CNVs. Accordingly, we propose that a dicentric chromosome intermediate is driving the rapid generation of copy number amplifications and sub-clonal heterogeneity observed in C. albicans clinical isolates.

Figure 6. Breakage-fusion-bridge model for the formation and resolution of complex CNVs.

DNA double-strand breaks (DSBs), recombination, and dicentric chromosome formation can drive complex CNV formation through successive breakage-fusion-bridge (BFB) cycles. (A) Two distinct long-repeat sequences, Repeat 124 (blue/dark blue arrows) and Repeat 127 (orange/dark orange arrows), are located on both Chr3 homologs (black and gray). Heterozygous SNVs indicated with an X on Chr3 homolog 2. (B) A DSB occurs telomere proximal to the left copy of Repeat 124 on homolog 1 of Chr3, resulting in the formation of a telomere-proximal acentric DNA fragment. DNA replication generates two sister chromatids with a truncated arm. Non-allelic homologous recombination (NAHR) between Repeat 124 on sister chromatids generates a dicentric chromosome. During cytokinesis, the dicentric chromosome undergoes bridge formation and breaks near the centromere due to the closure of the actomyosin ring, generating two asymmetric chromosome fragments, each with only one centromere. The longer, monocentric chromosome fragment is followed in this model. During the next cell cycle, DNA replication generates two sister chromatids that undergo NAHR between a distinct long inverted repeat (Repeat 127, orange arrows) on sister chromatids that generates a second dicentric chromosome. Subsequent BFB cycles can result in different outcomes depending on the environmental selection: (C) Complex CNV expansion in which successive rounds of the BFB cycle generate CNVs with higher copy numbers, or (D) NAHR with a repeat on the other homolog results in the resolution of the dicentric chromosome and isolates the CNV on a monocentric chromosome (top, e.g. AMS3092). Alternatively, break induced replication (BIR) could prime near CEN3 and continue to the Chr3L telomere, resulting in the loss of the complex CNV and homozygosis of Chr3L (bottom, e.g. AMS3093). Example stair-step copy numbers of genomic segments indicated below each schematic.

Figure 6.

Figure 6—figure supplement 1. Breakage-fusion-bridge model for the formation of asymmetric complex CNVs.

Figure 6—figure supplement 1.

CNVs with asymmetric stair-steps were associated with three distinct long inverted repeat sequences (e.g. Repeats X, Y, and Z) located on the same chromosome arm. A DNA double-strand break (DSB) occurs telomere proximal to the left copy of Repeat X (purple arrows) resulting in the formation of a telomere-proximal acentric DNA fragment. DNA replication generates two sister chromatids with a truncated arm. Non-allelic homologous recombination (NAHR) between Repeat X on sister chromatids generates a dicentric chromosome. During cytokinesis, the dicentric chromosome undergoes bridge formation and breaks between copies of Repeats Y and Z (blue and orange arrows), generating two asymmetric chromosome fragments, each with only one centromere. The longer, monocentric chromosome fragment is followed in this model. During the next cell cycle, DNA replication generates two sister chromatids that undergo NAHR between a distinct long inverted repeat (Repeat Y, blue arrows) on sister chromatids that generates a second dicentric chromosome. Again, following bridge formation, a DSB occurs between copies of Repeats X and Y (purple and blue arrows). DNA replication generates two sister chromatids that undergo NAHR between copies of Repeat X (purple arrows) on sister chromatids generating another dicentric chromosome. After a final DSB, NAHR with Repeat Z (dark orange) on the other homolog (gray) results in the resolution of the dicentric chromosome and isolates the CNV on a monocentric chromosome. Example asymmetric stair-step copy numbers of genomic segments indicated below model.

A limitation of this study is that the rate of complex CNV formation is not known, and it is also unclear whether these CNVs occur in the context of a host infection. Dicentric chromosomes have been identified in FLC-resistant clinical isolates of C. albicans (Selmecki et al., 2006), and suggest that BFB cycles may be possible in vivo. Additionally, CNVs with breakpoints at long-repeat sequences have been identified after passage of C. albicans through a mouse model of oropharyngeal candidiasis (in the absence of antifungal drugs) (Todd et al., 2019), and suggest that complex CNVs could occur in vivo. Importantly, there was very little fitness cost associated with maintaining a complex CNV on a monocentric chromosome (Figure 2), which underscores that these CNVs are well-tolerated in C. albicans. Ultimately, future studies using single-cell analyses (e.g. Lauer et al., 2018) are needed to determine the rate of complex CNV formation both in vitro and in vivo in the presence and absence of antifungal drugs.

Stair-step amplifications and fold-back inversions generated via BFB cycles are also observed in human cancers and developmental diseases (Cheng et al., 2016; Hermetz et al., 2014; Marotta et al., 2017). For example, amplification of oncogenes (EFGR, ERBB2, and MYC) occurs via BFB in up to ~70% of diverse tumor types (including breast, colorectal, lung, and liver cancers) (Marotta et al., 2017; Venkataram et al., 2016). While the exact molecular mechanisms of DNA repair during BFB in human cells remain under investigation (Cheng et al., 2016; Maciejowski et al., 2015; Marotta et al., 2012; Tanaka et al., 2007), fold-back inversions causing CNVs occur more frequently (13/21 vs. 3/21) at breakpoints of microhomology (2–8 bp) than at regions of longer homology (296–330 bp, 82–95%), such as LINE, SINE, and Alu elements that are present in thousands of copies per genome (Hermetz et al., 2014; Rodić and Burns, 2013). In comparison, the copy number breakpoints identified in C. albicans only occurred at long-repeat sequences (median 1513 bp) that shared high sequence identity (median 96.1%) that were predominantly present (10/12 repeats) in only two copies per genome and were always located on the same chromosome arm. Therefore, while C. albicans appears to have a strong preference for a BFB repair mechanism that involves homologous recombination within long-repeat sequences, the relatively low copy number of these repeat sequences in the C. albicans genome has enabled more precise mapping of the CNV breakpoints and fold-back inversions, whereas similar events remain challenging to resolve in the human genome.

Impact of CNVs on the mutational landscape

CNVs can dramatically alter population dynamics and mutational landscapes. The rate with which CNVs occur (~1.3×10−6 to ~1.5×10−6 per gene per cell division) is several orders of magnitude higher than the rate of SNVs (~0.33×10−9 per site per cell division) in the absence of selection (Lynch et al., 2008). In the presence of strong selection, for example nutrient limitation, the rate of CNV formation can be much higher and can rapidly drive clonal interference between unique CNV-containing lineages within the population (Lauer et al., 2018; Payen et al., 2014). Our single colony analyses support that formation of a dicentric chromosome can generate continuous cycles of genome instability and can increase population heterogeneity, which is likely to contribute to high rates of clonal interference.

We hypothesize that the complex CNVs identified here were selected due to the presence of genes that, when amplified, provide a fitness benefit in the presence of azole antifungal drugs (Selmecki et al., 2008). In support of this hypothesis, NPR2 (within the Chr3L CNV) results in increased azole resistance similar to acquisition of the CNV (Mount et al., 2018). Other genes like ERG11 and TAC1 (both within the i(5L) CNV) can confer copy number dependent increases in azole resistance (Selmecki et al., 2008).

In addition to the direct effect of gene amplification (i.e. more copies of the gene and its products), every cell that acquires a complex CNV also has an increased target size for the acquisition of rare, single nucleotide variants (Cone et al., 2017; Elde et al., 2012). CNVs therefore provide a potential for an increased likelihood of beneficial, drug-resistant point mutations during antifungal drug selection. This provides a good explanation for the utility of CNVs and aneuploidy as intermediates in the acquisition of more stable mutations (e.g. single nucleotide variants) (Elde et al., 2012; Ford et al., 2015; Roth and Andersson, 2004; Sun et al., 2009; Yona et al., 2012).

Importantly, many of the genes that are known to cause azole resistance contain SNVs that are recurrently found in drug-resistant isolates. For example, recurrent SNVs in MRR1, ERG11, FUR1, and FKS1 are found in drug-resistant isolates of C. albicans and interestingly, are recurrent in distantly related Candida species as well, including C. glabrata and C. auris (Flowers et al., 2015; Garcia-Effron et al., 2009; Hope et al., 2004; Morschhäuser et al., 2007; Muñoz et al., 2018; Perlin, 2011). Both MRR1 (Chr3R) and ERG11 (Chr5L), genes with recurrent SNVs known to cause azole resistance, were amplified in CNVs in this study, supporting the idea these CNVs can amplify genes that confer increased fitness in the presence of antifungal drugs and simultaneously increase the probability that a bona fide drug resistance allele will occur in those same genes.

LOH is also an important mechanism of acquired antifungal drug resistance, and exposure to azole antifungal drugs can increase the frequency of LOH in C. albicans (Bouchonville et al., 2009; Dunkel et al., 2008; Forche et al., 2011; Niimi et al., 2010). Heterozygous mutations in both MRR1 and FKS1 also can undergo LOH in the presence of antifungal drug selection for homozygosis of the beneficial allele (Dunkel et al., 2008; Niimi et al., 2010). Here, we obtained direct evidence that LOH can arise via dicentric chromosome formation and telomere-proximal chromosome loss, the same mechanism that yields complex CNVs. For example, new regions of LOH were identified during dicentric chromosome resolution (e.g. AMS4702 in Figure 1—figure supplement 3, and AMS3093 in Figure 5—figure supplement 2). These data further support that the long repeat sequences associated with CNV and LOH breakpoints are a major source of genome plasticity in the presence and absence of antifungal drug (Todd et al., 2019), and may underlie the variable mutation rates (hotspots) observed across an individual chromosome in other fungi as well (Lang and Murray, 2011).

Finally, we found that complex CNVs frequently occurred in regions of the genome that were already homozygous in the euploid progenitors (Figure 1—figure supplement 3). We propose that such homozygous regions are the result of prior rounds of complex CNV formation and subsequent LOH in some of these isolates. Other regions of long-track homozygous sequence are frequently observed in diverse clinical isolates of C. albicans (Ene et al., 2018; Ford et al., 2015; Hirakawa et al., 2015; Ropars et al., 2018; Todd et al., 2019) and may provide evidence that other complex CNVs resulting in LOH are occurring in some of these isolates as well, supporting the idea that these CNVs are transient in nature.

CNVs promote antifungal drug tolerance

Antifungal drug tolerance is the ability of a subpopulation of cells within a susceptible isolate to grow slowly at drug concentrations above the MIC50 (Berman and Krysan, 2020; Rosenberg et al., 2018). Mechanisms that contribute to the tolerance phenotype remain to be identified; however, genes involved in core stress pathways and cell wall/cell membrane biosynthesis appear to have an important role (Berman and Krysan, 2020; Cowen et al., 2015; Rosenberg et al., 2018). The complex CNVs detected here, together with what is known about drug responses in general, make it tempting to speculate on the specific genes that may be involved in the tolerance phenotype. These include genes encoding proteins involved in stress responses (CRZ1, HSP70, CGR1, ERO1, TPK1, ASR1, and PBS2) and cell wall/cell membrane integrity (CDR3, NCP1, ECM21, MNN23, RHB1 and KRE6). We propose that amplification of these genes within transient CNVs is one major route to increase cellular fitness and generate the population heterogeneity that underlies antifungal drug tolerance.

Impact of azole antifungals on centromere function and CNV formation

Whole chromosome missegregation resulting in aneuploidy is well-documented in C. albicans and other fungal pathogens (Forche et al., 2009; Forche et al., 2019; Gerstein et al., 2015; Janbon et al., 1998; Ngamskulrungroj et al., 2012; Poláková et al., 2009; Reedy et al., 2009; Rustchenko-Bulgac, 1991; Selmecki et al., 2006; Selmecki et al., 2009; Sionov et al., 2013; Yang et al., 2019). Many of these aneuploidies provide a selective benefit in vitro and in vivo, and in the presence and absence of antifungal drugs. Importantly, whole chromosome aneuploidy is likely to be induced (as well as selected) by drug exposure (Harrison et al., 2014); however, the mechanisms driving chromosome mis-segregation are not well characterized.

We find that the acquisition of complex CNVs involves the formation of a dicentric chromosome intermediate. The dicentric chromosomes are maintained in the presence of drug (Figure 5—figure supplement 3), whereas elimination of drug selection results in a ~13% increase in colonies that have lost the dicentric chromosome via subsequent recombination events that either isolate the complex CNV on a chromosome arm or revert to the euploid progenitor genotype (Figure 5A and B). Stabilization of dicentric chromosomes has been observed in humans, Drosophila melanogaster, Zea mays, and Schizosaccharomyces pombe due to centromere inactivation (Agudo et al., 2000; Earnshaw and Migeon, 1985; Han et al., 2006; Sato et al., 2012; Stimpson et al., 2012; Sullivan and Schwartz, 1995). Therefore, in addition to selection, we propose that the dicentric chromosome can be stabilized in the presence of azole drugs due to the depletion of Cse4/CENP-A, the centromere-specific histone H3. Importantly, Cse4/CENP-A, which is normally enriched at centromeric DNA, is depleted from the centromeres of C. albicans cells exposed to FLC (10 μg/ml), and contributes to an increased rate of chromosome mis-segregation (Brimacombe et al., 2019). Whether Cse4/CENP-A is actively removed from centromeric DNA in the presence of azole drugs is not known, but mammalian models suggest that CENP-A can be recruited away from the centromere to sites of DNA damage (Zeitlin et al., 2009). The presence of azole drugs may increase the likelihood that a dicentric chromosome is maintained and further increase the likelihood of recombination events that result in complex CNV formation. Therefore, determining the mechanistic link between antifungal drug treatment and centromere function is critical to understanding the genome instability that occurs during acquisition of drug resistance, including whole-chromosome aneuploidy, dicentric chromosome intermediates, and complex CNVs.

Conclusion

Complex CNVs are acquired rapidly during adaptation to azole antifungal drugs and cause an increase in drug resistance and tolerance. These CNVs are found across the genome, occur in diverse genetic backgrounds, and are all formed between a set of two distinct inverted repeat sequences that flank the amplified region. Evidence here provides support for a mechanism of CNV generation and resolution back to euploidy that is driven by successive BFB cycles involving a dicentric chromosome that repairs via homologous recombination between the repeat sequences. Furthermore, the cell-to-cell variability observed for clinical isolates during drug susceptibility assays may be due to the heterogeneity in the copy number of CNVs present within individual cells in a population as well as continued BFB cycles. Identification of CNVs in other pathogenic fungi, including the emerging multi-drug-resistant pathogen C. auris, further suggests that this mechanism of CNV formation may occur in diverse species. Together, these findings suggest a novel mechanism for transient CNV formation that increases the adaptive potential of fungal pathogens to antifungal drugs.

Materials and methods

Key resources table.

Reagent type
(species) or resource
Designation Source or reference Identifiers Additional
information
Strain, strain background (Candida albicans) SC5314 Hirakawa et al., 2015
(doi:10.1101/gr.174623.114)
RRID:SCR_013437
Strain, strain background (C. albicans) P75016 Hirakawa et al., 2015 (doi:10.1101/gr.174623.114)
Strain, strain background (C. albicans) P75063 Hirakawa et al., 2015
(doi:10.1101/gr.174623.114)
Strain, strain background (C. albicans) P78042 Hirakawa et al., 2015
(doi:10.1101/gr.174623.114)
Antibody Anti-Digoxigenin-AP Fab Fragments
(Polycolonal from sheep)
Roche 11093274910
RRID:AB_2734716
(1:5000)
Sequence-based reagent PCR Primers This study Supplementary file 5
Commercial assay or kit Illumina Nextera XT Library Kit Illumina 105032350
Commercial assay or kit Illumina Nextera XT Index Kit Illumina 105055294
Commercial assay or kit Illumina Nextera Flex DNA Kit Illumina 20018704
Commercial assay or kit Illumina Nextera DNA CN Index kit Illumina 20018707
Commercial assay or kit Blue Pippin 1.5% agarose gel dye-free cassette Sage Science 250 bp - 1.5 kb DNA size range collections, Marker R2 Target of 900 bp
Commercial assay or kit
Illumina MiSeq v2 Reagent Kit
Illumina
15033625
2 × 250 bp
Commercial assay or kit 1D Ligation Sequencing Kit Oxford Nanopore Technologies SQK-LSK108
Commercial assay or kit R9 FLO-Min106 spot-on flow cell Oxford Nanopore Technologies R9.4.1
Commercial assay or kit Ultra II End Repair/dA-Tailing Module New England Biolabs E7546S

Commercial assay or kit

Qubit dsDNA HS kit
Life Technologies
Q32854
Commercial assay or kit PCR DIG Probe Synthesis Kit Roche 11636090910
Commercial assay or kit
Agilent 2100 Bioanalyzer High Sensitivity DNA Reagents
Agilent Technologies
5067–4626
Commercial assay or kit
SacII restriction enzyme New England Biolabs R0157S
Chemical compound, drug Fluconazole (FLC) Alfa Aesar
J62015
Chemical compound, drug Miconazole Alfa Aesar AAJ6087206
Chemical compound, drug Itraconazole Alfa Aesar AAJ6639003
Chemical compound, drug Posaconazole MilliporeSigma 11-101-3331
Chemical compound, drug Ketoconazole Fisher Scientific AC455470010
Chemical compound, drug PMSF Milipore Sigma 10837091001
Software, algorithm Trimmomatic Bolger et al., 2014
(doi:10.1093/bioinformatics/btu170)
v0.33 RRID:SCR_011848
Software, algorithm BWA Li, 2013
(doi:10.1093/bioinformatics/btp324)
v0.7.12 RRID:SCR_010910
Software, algorithm Samtools Li et al., 2009
(doi:10.1093/bioinformatics/btp352)
v0.1.19 RRID:SCR_002105
Software, algorithm Genome Analysis Toolkit McKenna et al., 2010
(doi: 10.1101/gr.107524.110)
v3.4–46 RRID:SCR_001876
Software, algorithm Yeast Analysis Mapping Pipeline
Abbey et al., 2014
(doi: 10.1186/s13073-014-0100-8)
V1.0
Software, algorithm JMP Pro https://www.jmp.com V14.2.0
Software, algorithm ImageJ https://imagej.nih.gov/ij/?
v2.0.0-rc-30/1.49 s RRID:SCR_003070
Software, algorithm Integrative Genomics Viewer Thorvaldsdóttir et al., 2013 (doi:10.1093/bib/bbs017) v2.3.92 RRID:SCR_011793
Software, algorithm
R https://www.r-project.org
v3.5.2 RRID:SCR_001905
Software, algorithm
Candida Genome Database
http://Candidagenome.org
RRID:SCR_002036
Software, algorithm
NGMLR Sedlazeck et al., 2018
(doi: 10.1038/s41592-018-0001-7)
V0.2.7
Software, algorithm
Sniffles Sedlazeck et al., 2018
(doi: 10.1038/s41592-018-0001-7)
v1.0.11
Software, algorithm
SplitThreader Nattestad et al., 2016
(doi: 10.1101/087981) http://splitthreader.com
Software, algorithm
Ribbon http://genomeribbon.com

Yeast isolates and culture conditions

All isolates used in this study are described in Supplementary file 1. Isolates were stored at −80°C in 20% glycerol. Isolates were cultured at 30°C in YPAD medium (yeast extract, peptone, and 2% dextrose) supplemented with 40 µg/ml adenine and 80 µg/ml uridine. For Figure 5, isolates (AMS3050, AMS3053, AMS3054, AMS3052, and AMS3051) were grown at 30°C on YPAD agar plates (yeast extract, peptone, 2% dextrose, and 2% agar) for 48 hr. One single large and one single small colony (See ImageJ colony size analysis in Materials and methods) were selected and were re-plated for single colonies on YPAD agar plates for 48 hr at 30°C. From the large colony, a single large colony (AMS3092) was selected for WGS, antifungal drug susceptibility assays (MIC50), colony size analysis, and CHEF analysis. From the single small colony, both a small (AMS3094) and large (AMS3093) colony were selected for WGS, antifungal drug susceptibility assays (MIC50), colony size analysis, and CHEF analysis.

In vitro evolution experiment

FLC susceptible progenitor isolates were plated for single colonies onto YPAD medium and incubated for 48 hr at 30°C. Twelve single colonies were isolated from each progenitor (SC5314, P75016, P75063, and P78042) and grown to stationary phase in 5 ml liquid YPAD to generate 48 independent lineages. A 1:1000 cell dilution was made in YPAD medium containing 1 µg/ml FLC in deep-well 96-well plates. Plates were sealed with Breathe EASIER tape (Electron Microscopy Sciences) and placed in a humidified chamber for 72 hr at 30°C. Every 72 hr, cells were resuspended and transferred into fresh medium containing 1 µg/ml FLC to a final cell dilution of 1:1000. In total, 10 transfers were conducted. After the final transfer, cells were collected for storage at −80°C, genomic DNA isolation, and MIC analysis.

Microdilution minimum inhibitory concentration (MIC)

The MIC50 for each isolate was measured using a microwell broth dilution. Isolates were inoculated from frozen stocks into YPAD medium and grown for 16 hr at 30°C. Cells were diluted in fresh YPAD medium to a final OD600 of 0.01, and 10 µl of this dilution was inoculated into a 96-well plate containing 190 μl of a 0.5X dextrose YPAD medium with a twofold serial dilution of the antifungal drug or a no-drug control. Cells were incubated at 30°C in a humidified chamber and OD600 readings were taken at both 24 and 48 hr post inoculation. The MIC50 of each of the isolates was determined as the concentration of antifungal drug that decreased the OD600 by ≥50% of the no-drug control. Supra-MIC Growth (SMG) was calculated by taking the average 48 hr growth of the wells above the 24 hr MIC50 and dividing by the control well containing no drug (Rosenberg et al., 2018).

Growth curve analysis

Isolates were inoculated from frozen stocks into YPAD medium and grown for 16 hr at 30°C. Cells were diluted in fresh YPAD medium to a final OD600 of 0.01, and 10 µl of this dilution was inoculated into a 96-well plate containing 190 μl of a 1x dextrose YPAD medium with or without 1 μg/ml FLC. Cells were grown at 30°C in the BioTek Epoch with dual-orbital shaking (256 rpm) for 36 hr. OD600 readings were taken every 15 min and plotted in R (v3.5.2) using ggplot2. Each growth curve was conducted in biological triplicate in three separate experiments. Calculation of summary statistics of each growth curve was conducted using the R package Growthcurver using standard parameters (Supplementary file 3; Sprouffske and Wagner, 2016).

Contour-clamped homogenous electric field (CHEF) electrophoresis

Sample plugs were prepared as previously described (Selmecki et al., 2005). Briefly, cells were suspended in 300 µL 1.5% low-melt agarose (Bio-Rad) and digested with 1.2 mg Zymolyase (US Biological) at 37°C for 16 hr. Plugs were washed twice in 50 mM EDTA and treated with 0.2 mg/ml proteinase K (Alpha Azar) at 50°C for 48 hr. For samples digested with SacII, plugs were washed twice with 1x TE and incubated in 1 mM PMSF (Milipore Sigma) at room temperature for 30 min, washed twice with 1X TE and suspended in 1X CutSmart Buffer (New England Biolabs), and digested with 30 units of SacII (New England Biolabs) at 37°C for 24 hr. Chromosomes were separated in a 1% Megabase agarose gel (BioRad) in 0.5X TBE using the CHEF DRIII Pulsed Field Electrophoresis System. For whole chromosome separation, run conditions as follows: 60 s to 120 s switch, 6 V/cm, 120° angle for 36 hr followed by 120 s to 300 s switch, 4.5 V/cm, 120° angle for 12 hr. SacII digested chromosomes were separated as follows: 7 s to 100 s switch, 4.5 V/cm, 120° angle for 21 hr followed by 80 s to 400 s switch, 3.5 V/cm, 120° angle for 21 hr. CHEF gels were stained with ethidium bromide and imaged with the GelDock XR Imaging system (BioRad).

Southern blot hybridization

Chromosomes from CHEF gels were transferred to a BrightStar Plus nylon membrane (Invitrogen). Hybridization and detection of the DNA was conducted as previously described (Selmecki et al., 2005; Selmecki et al., 2008; Selmecki et al., 2009; Todd et al., 2019). Probes were generated through PCR incorporation of DIG-11-dUTP into target sequences following the manufacturer’s instructions (Roche). Primers used in this study are located in Supplementary file 5.

Gene Ontology (GO) analysis

GO analysis was conducted for all terms (process, function, and component) using the GO Term Finder from the Candida Genome Database (CGD accessed 03/03/2020, http://www.candidagenome.org/cgi-bin/GO/goTermFinder). All genes located within the complex CNVs were included in the analysis. Genes located in other aneuploid chromosomes (AMS4104 - Chr7; AMS4106 - Chr3, i(5L); AMS4444 - Chr3, Chr4R, Chr7R) were not included in the GO analysis. Terms were considered significantly enriched if p<0.05. GO enrichment was determined for all genes included in the complex CNVs, as well as for each CNV individually (Supplementary file 4).

ImageJ colony size analysis

All agar plates were imaged using the GelDock XR Imaging system (Bio-rad) using the same zoom and focal length. Images were exported as .png files, converted to eight-bit, and analyzed with Fiji (v2.0.0-rc-30/1.49 s) (Schindelin et al., 2012). An automatic threshold was set using the RenyiEntropy algorithm and the area of each particle was measured (Sezgin and Sankur, 2004). Colonies were considered small if their total area was less than two standard deviations below the mean colony size of the progenitor isolate, AMS3050 in the absence of miconazole. Colony size of each isolate, in the absence or presence of 20 μg/ml miconazole, was obtained from three individual agar plates (n > 113).

Illumina whole genome sequencing

Genomic DNA was isolated using a phenol-chloroform extraction as described previously (Selmecki et al., 2006). Libraries were prepared using either the Illumina Nextera XT DNA Library Preparation Kit or the Nextera DNA Flex Library Preparation Kit. Adaptor sequences and low-quality reads were trimmed using Trimmomatic (v0.33 LEADING:3 TRAILING:3 SLIDINGWINDOW:4:15 MINLEN:36 TOPHRED33) (Bolger et al., 2014). Trimmed reads were mapped to the C. albicans reference genome (A21-s02-m09-r08) from the Candida Genome Database (http://www.candidagenome.org/download/sequence/C_albicans_SC5314/Assembly21/archive/C_albicans_SC5314_version_A21-s02-m09-r08_chromosomes.fasta.gz). Reads were mapped using BWA-MEM (v0.7.12) with default parameters (Li, 2013). PCR duplicated reads were removed using Samtools (v0.1.19) (Li et al., 2009), and realigned around predicted indels using the Genome Analysis Toolkit (RealignerTargetCreator and IndelRealigner, v3.4–46) (McKenna et al., 2010). All Illumina data have been deposited in the National Center for Biotechnology Information Sequence Read Archive database under PRJNA613282.

Visualization of aneuploid chromosomes

Aneuploid chromosomes were visualized using the Yeast Analysis Mapping Pipeline (YMAP v1.0) (Abbey et al., 2014). Fastq files were uploaded to YMAP and read depth was determined and plotted as a function of chromosome position. Read depth was corrected for both GC-content and chromosome-end bias.

Read depth analysis plots

For each isolate, read depth for each position within the genome was calculated using samtools depth (--aa) (v0.1.19) (Li et al., 2009). A sliding window of 500 bp was used to calculate the average read depth over a given genome segment and normalized to the average read depth of the nuclear genome. Read depth analysis was visualized in R (v3.5.2) using ggplot2. Compiled read depth analysis found in Supplementary file 6.

CNV breakpoint detection

Chromosomes containing CNVs were detected using the Yeast Mapping Analysis Pipeline (YMAP v1.0 [Abbey et al., 2014]). Fastq files were uploaded and mapped to the SC5314 reference genome (A21-s02-m08-r09) with correction for GC-content and chromosome end bias. Estimated copy number breakpoints were detected using Aneufinder (v1.10.2), with a bin width of 42.5 kb (Bakker et al., 2016). The bins containing estimated breakpoints were identified for further analysis. Estimated breakpoints located at the MRSs were not analyzed due to the repetitive nature and poor mapping of these genomic regions. To further refine copy number breaks, fastq files were aligned to the SC5314 reference genome (see above) and read depth was determined for each nucleotide in the genome (samtools depth --aa, v0.1.19). Read depth was normalized to the mean read depth of the nuclear genome using R (v3.5.2). The 42.5 kb windows containing estimated copy number breakpoints as determined by Aneufinder were further subdivided into 500 bp windows. The mean normalized read depth was determined for these 500 bp windows and a rolling mean of every two consecutive 500 bp windows was determined using R. Copy number breakpoints were identified if 75% of four consecutive 500 bp windows had a mean normalized read depth that deviated from the mean nuclear genome read depth by more than 25% (Ford et al., 2015). Boundaries were further confirmed by visual inspection in Integrative Genomics Viewer (IGV v2.3.92) (Thorvaldsdóttir et al., 2013). CNV breakpoint positions were compared to the list of long repeat sequences found across the C. albicans genome described in Supplementary file 2 of Todd et al., 2019 and breakpoints were assigned a repeat name if they fell within 2 kb of a long repeat sequence (Todd et al., 2019).

Copy number detection of complex CNVs

For each isolate, read depth for each position within the genome was calculated using samtools depth (--aa) (v0.1.19) (Li et al., 2009). A sliding window of 500 bp was used to calculate the average read depth over a given genome segment and normalized to the average read depth of the nuclear genome. Each segment of the chromosome containing a complex CNV (high copy number central region and all lower copy number flanking sequences, including the disomic regions) was then assigned a copy number using the average normalized read depth of the 500 bp windows located within that given CNV segment multiplied by two for the diploid genome. If a 500 bp window contained a known repeat sequence (including those associated with copy number breakpoints) that window was excluded from analysis due to read mapping errors that occur in repetitive sequences. Likewise, telomere proximal regions were excluded from analysis. Telomere proximal regions were determined as the start or end of each chromosome to the first confirmed, non-repetitive genome feature as previously described (Ene et al., 2018; Hirakawa et al., 2015; Todd et al., 2019). The read depth for each segment was then normalized to the average read depth of the disomic segments of the chromosome to normalize for chromosome copy number, and rounded to the nearest integer for the final copy number. The copy number data, as well as read depth summary statistics for each chromosome segment are found in Supplementary file 6. Data were analyzed and summary statistics were generated using JMP Pro (v14.2.0).

Allele ratio analysis

Heterozygous positions were determined using the Genomics Analysis Toolkit’s HaplotypeCaller (v3.7–0-gcfed67) with a standard minimum confidence threshold of phred30 (-stand_call_conf 30) (McKenna et al., 2010). Variants were analyzed if the read depth was >2 and the variant was sequenced on both the forward and reverse strand. Additional filtering of SNVs included the removal of ancestral homozygous positions, homozygous SNVs that were maintained within the progenitor and all evolved isolates and not the reference SC5314. SNVs contained within long-repeat sequences were also removed due to the mapping errors with short read sequencing in repetitive regions.

Variant calling

De novo variant detection was conducted using aligned, sorted BAM files that were converted to mpileup files using Samtools (samtools mpileup) (Li et al., 2009). VCFs were generated using Varscan (V2.3) (Koboldt et al., 2012). Called variants were filtered under the following conditions: 1) All SNVs shared by all progeny are removed and are assumed to be parentally derived, 2) a read depth of <5, and 3) a percent alternative allele of <0.2.

Oxford Nanopore Technology MinION sequencing and de novo alignment

Two identical libraries of AMS3051 were constructed using the 1D Ligation Sequencing Kit (SQK-LSK108) from Oxford Nanopore following manufacturers protocol with slight modification. Briefly, end repair and dA-tailing was performed following New England Biolabs protocol for the Ultra II End-prep reaction (NEB E7546S) with a 30 min incubation at 20°C followed by a 30 min incubation at 65°C. The DNA was then purified using 1.8x Ampure beads (Agencourt). Adapter ligation was allowed to incubate at room temperature for 30 min followed by a 0.6x Ampure (Agencourt) bead cleanup. Data was generated using the R9 FLO-MIN106 spot-on flow cell. Calibration of the flow cell indicated 1171 active pores across the four mux groups. The library was loaded following manufacturers recommendation and sequencing was allowed to proceed for 24 hr before loading with the second library. Sequencing continued for another 24 hr before the sequencing run was terminated. A total of 910753 reads were obtained with a mean read length of 2210 bp (minimum 5 bp and maximum 352979 bp). Average theoretical coverage was 129.9x assuming a haploid genome size of 15.5 Mb.

Visualization of Oxford Nanopore Technology MinION sequencing data

Oxford Nanopore minion Fastq files from AMS3051 were aligned to the SC5314 reference genome (A21-s02-m09-r08) using NGMLR (-x ont, v0.2.7) (Sedlazeck et al., 2018). The resulting SAM file was converted to a BAM file using samtools view (-S –b) and was sorted and indexed using samtools sort and samtools index, respectively (v0.1.19) (Li et al., 2009). Structural variant detection was conducted using Sniffles (v1.0.11) (Sedlazeck et al., 2018) and the average binned (10 kb) read coverage was determined using Copycat (https://github.com/MariaNattestad/copycat). Structural variants were identified using SplitThreader (http://splitthreader.com). Individual discordant reads were identified and visualized using Ribbon with a minimum alignment length of 1 kb (Nattestad et al., 2016).

Acknowledgements

We thank Curtis Focht from our lab for assistance developing multiple bioinformatics pipelines, and Hung-ji Tsai, Laura Burrack, and Dana Davis for helpful discussions and feedback on the manuscript. We are grateful to Leah Cowen and her lab for the isolates detailed in Figure 3. Support for this research was provided by NIH grant R01 AI143689, NE Established Program to Stimulate Competitive Research (EPSCoR) First Award, NE Department of Health and Human Services (LB506-2017-55) award, and NIH-NCRR COBRE grant P20RR018788 sub-award. The sequencing datasets generated during this study are available in the Sequence Read Archive repository under BioProject PRJNA613282.

Funding Statement

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Contributor Information

Anna Selmecki, Email: selmecki@umn.edu.

Kevin J Verstrepen, VIB-KU Leuven Center for Microbiology, Belgium.

Patricia J Wittkopp, University of Michigan, United States.

Funding Information

This paper was supported by the following grants:

  • National Institute of Allergy and Infectious Diseases AI143689 to Anna Selmecki.

  • Nebraska Department of Health and Human Services LB506-2017-55 to Anna Selmecki.

  • Nebraska Established Program to Stimulate Competitive Research First Award to Anna Selmecki.

  • National Center for Research Resources COBRE P20RR018788 sub-award to Anna Selmecki.

Additional information

Competing interests

No competing interests declared.

Author contributions

Conceptualization, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing - original draft, Writing - review and editing.

Conceptualization, Data curation, Formal analysis, Supervision, Funding acquisition, Visualization, Methodology, Writing - original draft, Writing - review and editing.

Additional files

Supplementary file 1. Strains used in this study.
elife-58349-supp1.xlsx (11.1KB, xlsx)
Supplementary file 2. Features of long-repeat sequences identified at copy number breakpoints of complex CNVs.
elife-58349-supp2.xlsx (16.5KB, xlsx)
Supplementary file 3. Growth curve raw data and analysis.
elife-58349-supp3.xlsx (74.9KB, xlsx)
Supplementary file 4. Coding sequences within complex CNVs.
elife-58349-supp4.xlsx (191.3KB, xlsx)
Supplementary file 5. Primers used in this study.
elife-58349-supp5.xlsx (32.3KB, xlsx)
Supplementary file 6. Summary of complex CNV features.
elife-58349-supp6.xlsx (79.8KB, xlsx)
Transparent reporting form

Data availability

All genome sequencing data have been deposited in the Sequence Read Archive under BioProject PRJNA613282 All data analyzed during this study are included in the manuscript and supporting files. The source data file is provided for Figure 2.

The following dataset was generated:

Todd RT, Selmecki A. 2020. Complex copy number variants in Candida albicans. NCBI BioProject. PRJNA613282

The following previously published dataset was used:

Mount HO, Revie NM, Todd RT, Anstett K, Collins C, Costanzo M, Boone C, Robbins N, Selmecki A, Cowen LE. 2018. Candida albicans genome sequencing. NCBI BioProject. PRJNA323475

References

  1. Abbey DA, Funt J, Lurie-Weinberger MN, Thompson DA, Regev A, Myers CL, Berman J. YMAP: a pipeline for visualization of copy number variation and loss of heterozygosity in eukaryotic pathogens. Genome Medicine. 2014;6:100. doi: 10.1186/s13073-014-0100-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Adamo GM, Lotti M, Tamás MJ, Brocca S. Amplification of the CUP1 gene is associated with evolution of copper tolerance in Saccharomyces cerevisiae. Microbiology. 2012;158:2325–2335. doi: 10.1099/mic.0.058024-0. [DOI] [PubMed] [Google Scholar]
  3. Adler M, Anjum M, Berg OG, Andersson DI, Sandegren L. High fitness costs and instability of gene duplications reduce rates of evolution of new genes by duplication-divergence mechanisms. Molecular Biology and Evolution. 2014;31:1526–1535. doi: 10.1093/molbev/msu111. [DOI] [PubMed] [Google Scholar]
  4. Agudo M, Abad JP, Molina I, Losada A, Ripoll P, Villasante A. A dicentric chromosome of Drosophila melanogaster showing alternate centromere inactivation. Chromosoma. 2000;109:190–196. doi: 10.1007/s004120050427. [DOI] [PubMed] [Google Scholar]
  5. Alby K, Schaefer D, Bennett RJ. Homothallic and heterothallic mating in the opportunistic pathogen candida albicans. Nature. 2009;460:890–893. doi: 10.1038/nature08252. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Anderson RP, Roth JR. Tandem genetic duplications in phage and Bacteria. Annual Review of Microbiology. 1977;31:473–505. doi: 10.1146/annurev.mi.31.100177.002353. [DOI] [PubMed] [Google Scholar]
  7. Bagel S, Hüllen V, Wiedemann B, Heisig P. Impact of gyrA and parCMutations on quinolone resistance, doubling time, and supercoiling degree of Escherichia coli. Antimicrobial Agents and Chemotherapy. 1999;43:868–875. doi: 10.1128/AAC.43.4.868. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bakker B, Taudt A, Belderbos ME, Porubsky D, Spierings DC, de Jong TV, Halsema N, Kazemier HG, Hoekstra-Wakker K, Bradley A, de Bont ES, van den Berg A, Guryev V, Lansdorp PM, Colomé-Tatché M, Foijer F. Single-cell sequencing reveals karyotype heterogeneity in murine and human malignancies. Genome Biology. 2016;17:115. doi: 10.1186/s13059-016-0971-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Basra P, Alsaadi A, Bernal-Astrain G, O'Sullivan ML, Hazlett B, Clarke LM, Schoenrock A, Pitre S, Wong A. Fitness tradeoffs of antibiotic resistance in extraintestinal pathogenic Escherichia coli. Genome Biology and Evolution. 2018;10:667–679. doi: 10.1093/gbe/evy030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Bayer A, Brennan G, Geballe AP. Adaptation by copy number variation in monopartite viruses. Current Opinion in Virology. 2018;33:7–12. doi: 10.1016/j.coviro.2018.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Ben-Ami R, Zimmerman O, Finn T, Amit S, Novikov A, Wertheimer N, Lurie-Weinberger M, Berman J. Heteroresistance to fluconazole is a continuously distributed phenotype among candida glabrata clinical strains associated with in vivo persistence. mBio. 2016;7:e00655-16. doi: 10.1128/mBio.00655-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Berman J, Krysan DJ. Drug resistance and tolerance in fungi. Nature Reviews Microbiology. 2020;18:319–331. doi: 10.1038/s41579-019-0322-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Beroukhim R, Mermel CH, Porter D, Wei G, Raychaudhuri S, Donovan J, Barretina J, Boehm JS, Dobson J, Urashima M, Mc Henry KT, Pinchback RM, Ligon AH, Cho YJ, Haery L, Greulich H, Reich M, Winckler W, Lawrence MS, Weir BA, Tanaka KE, Chiang DY, Bass AJ, Loo A, Hoffman C, Prensner J, Liefeld T, Gao Q, Yecies D, Signoretti S, Maher E, Kaye FJ, Sasaki H, Tepper JE, Fletcher JA, Tabernero J, Baselga J, Tsao MS, Demichelis F, Rubin MA, Janne PA, Daly MJ, Nucera C, Levine RL, Ebert BL, Gabriel S, Rustgi AK, Antonescu CR, Ladanyi M, Letai A, Garraway LA, Loda M, Beer DG, True LD, Okamoto A, Pomeroy SL, Singer S, Golub TR, Lander ES, Getz G, Sellers WR, Meyerson M. The landscape of somatic copy-number alteration across human cancers. Nature. 2010;463:899–905. doi: 10.1038/nature08822. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Bicanic T, Muzoora C, Brouwer AE, Meintjes G, Longley N, Taseera K, Rebe K, Loyse A, Jarvis J, Bekker LG, Wood R, Limmathurotsakul D, Chierakul W, Stepniewska K, White NJ, Jaffar S, Harrison TS. Independent association between rate of clearance of infection and clinical outcome of HIV-associated cryptococcal meningitis: analysis of a combined cohort of 262 patients. Clinical Infectious Diseases. 2009;49:702–709. doi: 10.1086/604716. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for illumina sequence data. Bioinformatics. 2014;30:2114–2120. doi: 10.1093/bioinformatics/btu170. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Bouchonville K, Forche A, Tang KE, Selmecki A, Berman J. Aneuploid chromosomes are highly unstable during DNA transformation of candida albicans. Eukaryotic Cell. 2009;8:1554–1566. doi: 10.1128/EC.00209-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Bravo Ruiz G, Ross ZK, Holmes E, Schelenz S, Gow NAR, Lorenz A. Rapid and extensive karyotype diversification in haploid clinical candida auris isolates. Current Genetics. 2019;65:1217–1228. doi: 10.1007/s00294-019-00976-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Brewer BJ, Payen C, Raghuraman MK, Dunham MJ. Origin-dependent inverted-repeat amplification: a replication-based model for generating palindromic amplicons. PLOS Genetics. 2011;7:e1002016. doi: 10.1371/journal.pgen.1002016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Brewer BJ, Payen C, Di Rienzi SC, Higgins MM, Ong G, Dunham MJ, Raghuraman MK. Origin-Dependent Inverted-Repeat amplification: tests of a model for inverted DNA amplification. PLOS Genetics. 2015;11:e1005699. doi: 10.1371/journal.pgen.1005699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Brimacombe CA, Burke JE, Parsa JY, Catania S, O'Meara TR, Witchley JN, Burrack LS, Madhani HD, Noble SM. A natural histone H2A variant lacking the Bub1 phosphorylation site and regulated depletion of centromeric histone CENP-A foster evolvability in Candida Albicans. PLOS Biology. 2019;17:e3000331. doi: 10.1371/journal.pbio.3000331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Brown CJ, Todd KM, Rosenzweig RF. Multiple duplications of yeast hexose transport genes in response to selection in a glucose-limited environment. Molecular Biology and Evolution. 1998;15:931–942. doi: 10.1093/oxfordjournals.molbev.a026009. [DOI] [PubMed] [Google Scholar]
  22. Brown GD, Netea MG. Exciting developments in the immunology of fungal infections. Cell Host & Microbe. 2012;11:422–424. doi: 10.1016/j.chom.2012.04.010. [DOI] [PubMed] [Google Scholar]
  23. Chen SC, Sorrell TC. Antifungal agents. Medical Journal of Australia. 2007;187:404–409. doi: 10.5694/j.1326-5377.2007.tb01313.x. [DOI] [PubMed] [Google Scholar]
  24. Cheng C, Zhou Y, Li H, Xiong T, Li S, Bi Y, Kong P, Wang F, Cui H, Li Y, Fang X, Yan T, Li Y, Wang J, Yang B, Zhang L, Jia Z, Song B, Hu X, Yang J, Qiu H, Zhang G, Liu J, Xu E, Shi R, Zhang Y, Liu H, He C, Zhao Z, Qian Y, Rong R, Han Z, Zhang Y, Luo W, Wang J, Peng S, Yang X, Li X, Li L, Fang H, Liu X, Ma L, Chen Y, Guo S, Chen X, Xi Y, Li G, Liang J, Yang X, Guo J, Jia J, Li Q, Cheng X, Zhan Q, Cui Y. Whole-Genome sequencing reveals diverse models of structural variations in esophageal squamous cell carcinoma. The American Journal of Human Genetics. 2016;98:256–274. doi: 10.1016/j.ajhg.2015.12.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Chibana H, Iwaguchi S, Homma M, Chindamporn A, Nakagawa Y, Tanaka K. Diversity of tandemly repetitive sequences due to short periodic repetitions in the chromosomes of candida albicans. Journal of Bacteriology. 1994;176:3851–3858. doi: 10.1128/JB.176.13.3851-3858.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Chibana H, Beckerman JL, Magee PT. Fine-resolution physical mapping of genomic diversity in candida albicans. Genome Research. 2000;10:1865–1877. doi: 10.1101/gr.148600. [DOI] [PubMed] [Google Scholar]
  27. Chindamporn A, Nakagawa Y, Mizuguchi I, Chibana H, Doi M, Tanaka K. Repetitive sequences (RPSs) in the chromosomes of Candida Albicans are sandwiched between two novel stretches, HOK and RB2, common to each chromosome. Microbiology. 1998;144:849–857. doi: 10.1099/00221287-144-4-849. [DOI] [PubMed] [Google Scholar]
  28. Chow EW, Morrow CA, Djordjevic JT, Wood IA, Fraser JA. Microevolution of cryptococcus neoformans driven by massive tandem gene amplification. Molecular Biology and Evolution. 2012;29:1987–2000. doi: 10.1093/molbev/mss066. [DOI] [PubMed] [Google Scholar]
  29. Colombo AL, Guimarães T, Sukienik T, Pasqualotto AC, Andreotti R, Queiroz-Telles F, Nouér SA, Nucci M. Prognostic factors and historical trends in the epidemiology of candidemia in critically ill patients: an analysis of five multicenter studies sequentially conducted over a 9-year period. Intensive Care Medicine. 2014;40:1489–1498. doi: 10.1007/s00134-014-3400-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Cone KR, Kronenberg ZN, Yandell M, Elde NC. Emergence of a viral RNA polymerase variant during gene copy number amplification promotes rapid evolution of vaccinia virus. Journal of Virology. 2017;91:e01428-16. doi: 10.1128/JVI.01428-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Coste AT, Karababa M, Ischer F, Bille J, Sanglard D. TAC1, transcriptional activator of CDR genes, is a new transcription factor involved in the regulation of Candida Albicans ABC transporters CDR1 and CDR2. Eukaryotic Cell. 2004;3:1639–1652. doi: 10.1128/EC.3.6.1639-1652.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Coste A, Turner V, Ischer F, Morschhäuser J, Forche A, Selmecki A, Berman J, Bille J, Sanglard D. 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. 2006;172:2139–2156. doi: 10.1534/genetics.105.054767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Coste A, Selmecki A, Forche A, Diogo D, Bougnoux ME, d'Enfert C, Berman J, Sanglard D. Genotypic evolution of azole resistance mechanisms in sequential Candida Albicans isolates. Eukaryotic Cell. 2007;6:1889–1904. doi: 10.1128/EC.00151-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Côte P, Hogues H, Whiteway M. Transcriptional analysis of the candida albicans cell cycle. Molecular Biology of the Cell. 2009;20:3363–3373. doi: 10.1091/mbc.e09-03-0210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Cowen LE, Sanglard D, Howard SJ, Rogers PD, Perlin DS. Mechanisms of antifungal drug resistance. Cold Spring Harbor Perspectives in Medicine. 2015;5:a019752. doi: 10.1101/cshperspect.a019752. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Croll D, Zala M, McDonald BA. Breakage-fusion-bridge cycles and large insertions contribute to the rapid evolution of accessory chromosomes in a fungal pathogen. PLOS Genetics. 2013;9:e1003567. doi: 10.1371/journal.pgen.1003567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Croll D, McDonald BA. The accessory genome as a cradle for adaptive evolution in pathogens. PLOS Pathogens. 2012;8:e1002608. doi: 10.1371/journal.ppat.1002608. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Delarze E, Sanglard D. Defining the frontiers between antifungal resistance, tolerance and the concept of persistence. Drug Resistance Updates. 2015;23:12–19. doi: 10.1016/j.drup.2015.10.001. [DOI] [PubMed] [Google Scholar]
  39. Deng SK, Yin Y, Petes TD, Symington LS. Mre11-Sae2 and RPA collaborate to prevent palindromic gene amplification. Molecular Cell. 2015;60:500–508. doi: 10.1016/j.molcel.2015.09.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Dulmage KA, Darnell CL, Vreugdenhil A, Schmid AK. Copy number variation is associated with gene expression change in archaea. Microbial Genomics. 2018;4:e000210. doi: 10.1099/mgen.0.000210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Dunkel N, Blaß J, Rogers PD, Morschhäuser J. 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. Molecular Microbiology. 2008;69:827–840. doi: 10.1111/j.1365-2958.2008.06309.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Earnshaw WC, Migeon BR. Three related centromere proteins are absent from the inactive centromere of a stable isodicentric chromosome. Chromosoma. 1985;92:290–296. doi: 10.1007/BF00329812. [DOI] [PubMed] [Google Scholar]
  43. Elde NC, Child SJ, Eickbush MT, Kitzman JO, Rogers KS, Shendure J, Geballe AP, Malik HS. Poxviruses deploy genomic accordions to adapt rapidly against host antiviral defenses. Cell. 2012;150:831–841. doi: 10.1016/j.cell.2012.05.049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Ene IV, Farrer RA, Hirakawa MP, Agwamba K, Cuomo CA, Bennett RJ. Global analysis of mutations driving microevolution of a heterozygous diploid fungal pathogen. PNAS. 2018;115:E8688–E8697. doi: 10.1073/pnas.1806002115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Enjalbert B, Smith DA, Cornell MJ, Alam I, Nicholls S, Brown AJ, Quinn J. Role of the Hog1 stress-activated protein kinase in the global transcriptional response to stress in the fungal pathogen candida albicans. Molecular Biology of the Cell. 2006;17:1018–1032. doi: 10.1091/mbc.e05-06-0501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Eschrich D, Buchhaupt M, Kötter P, Entian KD. Nep1p (Emg1p), a novel protein conserved in eukaryotes and archaea, is involved in ribosome biogenesis. Current Genetics. 2002;40:326–338. doi: 10.1007/s00294-001-0269-4. [DOI] [PubMed] [Google Scholar]
  47. Felton T, Troke PF, Hope WW. Tissue penetration of antifungal agents. Clinical Microbiology Reviews. 2014;27:68–88. doi: 10.1128/CMR.00046-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Finn KJ, Li JJ. Single-stranded annealing induced by re-initiation of replication origins provides a novel and efficient mechanism for generating copy number expansion via non-allelic homologous recombination. PLOS Genetics. 2013;9:e1003192. doi: 10.1371/journal.pgen.1003192. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Flowers SA, Colón B, Whaley SG, Schuler MA, Rogers PD. Contribution of Clinically Derived Mutations in ERG11 to Azole Resistance in Candida albicans. Antimicrobial Agents and Chemotherapy. 2015;59:450–460. doi: 10.1128/AAC.03470-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Forche A, Alby K, Schaefer D, Johnson AD, Berman J, Bennett RJ. The parasexual cycle in Candida Albicans provides an alternative pathway to meiosis for the formation of recombinant strains. PLOS Biology. 2008;6:e110. doi: 10.1371/journal.pbio.0060110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Forche A, Magee PT, Selmecki A, Berman J, May G. Evolution in Candida Albicans populations during a single passage through a mouse host. Genetics. 2009;182:799–811. doi: 10.1534/genetics.109.103325. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Forche A, Abbey D, Pisithkul T, Weinzierl MA, Ringstrom T, Bruck D, Petersen K, Berman J. Stress alters rates and types of loss of heterozygosity in candida albicans. mBio. 2011;2:e00129-11. doi: 10.1128/mBio.00129-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Forche A, Solis NV, Swidergall M, Thomas R, Guyer A, Beach A, Cromie GA, Le GT, Lowell E, Pavelka N, Berman J, Dudley AM, Selmecki A, Filler SG. Selection of candida albicans trisomy during oropharyngeal infection results in a commensal-like phenotype. PLOS Genetics. 2019;15:e1008137. doi: 10.1371/journal.pgen.1008137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Ford CB, Funt JM, Abbey D, Issi L, Guiducci C, Martinez DA, Delorey T, Li BY, White TC, Cuomo C, Rao RP, Berman J, Thompson DA, Regev A. The evolution of drug resistance in clinical isolates of candida albicans. eLife. 2015;4:e00662. doi: 10.7554/eLife.00662. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Garcia-Effron G, Lee S, Park S, Cleary JD, Perlin DS. Effect of candida glabrata FKS1 and FKS2 mutations on echinocandin sensitivity and kinetics of 1,3-β-d-Glucan synthase: implication for the existing susceptibility breakpoint. Antimicrobial Agents and Chemotherapy. 2009;53:3690–3699. doi: 10.1128/AAC.00443-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Garnaud C, García-Oliver E, Wang Y, Maubon D, Bailly S, Despinasse Q, Champleboux M, Govin J, Cornet M. The rim pathway mediates antifungal tolerance in Candida Albicans through newly identified Rim101 transcriptional targets, including Hsp90 and Ipt1. Antimicrobial Agents and Chemotherapy. 2018;62:e01785-17. doi: 10.1128/AAC.01785-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Gerstein AC, Fu MS, Mukaremera L, Li Z, Ormerod KL, Fraser JA, Berman J, Nielsen K. Polyploid titan cells produce haploid and aneuploid progeny to promote stress adaptation. mBio. 2015;6:e01340. doi: 10.1128/mBio.01340-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Gerstein AC, Lim H, Berman J, Hickman MA. Ploidy tug-of-war: evolutionary and genetic environments influence the rate of ploidy drive in a human fungal pathogen. Evolution. 2017;71:1025–1038. doi: 10.1111/evo.13205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Ghannoum MA, Rice LB. Antifungal agents: mode of action, mechanisms of resistance, and correlation of these mechanisms with bacterial resistance. Clinical Microbiology Reviews. 1999;12:501–517. doi: 10.1128/CMR.12.4.501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Green BM, Finn KJ, Li JJ. Loss of DNA replication control is a potent inducer of gene amplification. Science. 2010;329:943–946. doi: 10.1126/science.1190966. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Gresham D, Desai MM, Tucker CM, Jenq HT, Pai DA, Ward A, DeSevo CG, Botstein D, Dunham MJ. The repertoire and dynamics of evolutionary adaptations to controlled nutrient-limited environments in yeast. PLOS Genetics. 2008;4:e1000303. doi: 10.1371/journal.pgen.1000303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Gresham D, Usaite R, Germann SM, Lisby M, Botstein D, Regenberg B. Adaptation to diverse nitrogen-limited environments by deletion or extrachromosomal element formation of the GAP1 locus. PNAS. 2010;107:18551–18556. doi: 10.1073/pnas.1014023107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Haber JE, Debatisse M. Gene amplification: yeast takes a turn. Cell. 2006;125:1237–1240. doi: 10.1016/j.cell.2006.06.012. [DOI] [PubMed] [Google Scholar]
  64. Han F, Lamb JC, Birchler JA. High frequency of centromere inactivation resulting in stable dicentric chromosomes of maize. PNAS. 2006;103:3238–3243. doi: 10.1073/pnas.0509650103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Harrison BD, Hashemi J, Bibi M, Pulver R, Bavli D, Nahmias Y, Wellington M, Sapiro G, Berman J. A tetraploid intermediate precedes aneuploid formation in yeasts exposed to fluconazole. PLOS Biology. 2014;12:e1001815. doi: 10.1371/journal.pbio.1001815. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Hastings PJ, Ira G, Lupski JR. A microhomology-mediated break-induced replication model for the origin of human copy number variation. PLOS Genetics. 2009;5:e1000327. doi: 10.1371/journal.pgen.1000327. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Heitzer E, Ulz P, Geigl JB, Speicher MR. Non-invasive detection of genome-wide somatic copy number alterations by liquid biopsies. Molecular Oncology. 2016;10:494–502. doi: 10.1016/j.molonc.2015.12.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Hermetz KE, Newman S, Conneely KN, Martin CL, Ballif BC, Shaffer LG, Cody JD, Rudd MK. Large inverted duplications in the human genome form via a fold-back mechanism. PLOS Genetics. 2014;10:e1004139. doi: 10.1371/journal.pgen.1004139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Hickman MA, Zeng G, Forche A, Hirakawa MP, Abbey D, Harrison BD, Wang YM, Su CH, Bennett RJ, Wang Y, Berman J. The 'obligate diploid' Candida albicans forms mating-competent haploids. Nature. 2013;494:55–59. doi: 10.1038/nature11865. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Hieronymus H, Murali R, Tin A, Yadav K, Abida W, Moller H, Berney D, Scher H, Carver B, Scardino P, Schultz N, Taylor B, Vickers A, Cuzick J, Sawyers CL. Tumor copy number alteration burden is a pan-cancer prognostic factor associated with recurrence and death. eLife. 2018;7:e37294. doi: 10.7554/eLife.37294. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Hirakawa MP, Martinez DA, Sakthikumar S, Anderson MZ, Berlin A, Gujja S, Zeng Q, Zisson E, Wang JM, Greenberg JM, Berman J, Bennett RJ, Cuomo CA. Genetic and phenotypic intra-species variation in candida albicans. Genome Research. 2015;25:413–425. doi: 10.1101/gr.174623.114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Hope WW, Tabernero L, Denning DW, Anderson MJ. Molecular mechanisms of primary resistance to flucytosine in Candida Albicans. Antimicrobial Agents and Chemotherapy. 2004;48:4377–4386. doi: 10.1128/AAC.48.11.4377-4386.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Hull RM, Cruz C, Jack CV, Houseley J. Environmental change drives accelerated adaptation through stimulated copy number variation. PLOS Biology. 2017;15:e2001333. doi: 10.1371/journal.pbio.2001333. [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Hull RM, King M, Pizza G, Krueger F, Vergara X, Houseley J. Transcription-induced formation of extrachromosomal DNA during yeast ageing. PLOS Biology. 2019;17:e3000471. doi: 10.1371/journal.pbio.3000471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Hull CM, Johnson AD. Identification of a mating type-like locus in the asexual pathogenic yeast candida albicans. Science. 1999;285:1271–1275. doi: 10.1126/science.285.5431.1271. [DOI] [PubMed] [Google Scholar]
  76. Hwang S, Gustafsson HT, O'Sullivan C, Bisceglia G, Huang X, Klose C, Schevchenko A, Dickson RC, Cavaliere P, Dephoure N, Torres EM. Serine-Dependent sphingolipid synthesis is a metabolic liability of aneuploid cells. Cell Reports. 2017;21:3807–3818. doi: 10.1016/j.celrep.2017.11.103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Janbon G, Sherman F, Rustchenko E. Monosomy of a specific chromosome determines L-sorbose utilization: a novel regulatory mechanism in candida albicans. PNAS. 1998;95:5150–5155. doi: 10.1073/pnas.95.9.5150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Koboldt DC, Zhang Q, Larson DE, Shen D, McLellan MD, Lin L, Miller CA, Mardis ER, Ding L, Wilson RK. VarScan 2: somatic mutation and copy number alteration discovery in Cancer by exome sequencing. Genome Research. 2012;22:568–576. doi: 10.1101/gr.129684.111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  79. Lang GI, Parsons L, Gammie AE. Mutation rates, Spectra, and genome-wide distribution of spontaneous mutations in mismatch repair deficient yeast. G3: Genes, Genomes, Genetics. 2013;3:1453–1465. doi: 10.1534/g3.113.006429. [DOI] [PMC free article] [PubMed] [Google Scholar]
  80. Lang GI, Murray AW. Mutation rates across budding yeast chromosome VI are correlated with replication timing. Genome Biology and Evolution. 2011;3:799–811. doi: 10.1093/gbe/evr054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  81. Lauer S, Avecilla G, Spealman P, Sethia G, Brandt N, Levy SF, Gresham D. Single-cell copy number variant detection reveals the dynamics and diversity of adaptation. PLOS Biology. 2018;16:e3000069. doi: 10.1371/journal.pbio.3000069. [DOI] [PMC free article] [PubMed] [Google Scholar]
  82. Lephart PR, Magee PT. Effect of the major repeat sequence on mitotic recombination in Candida Albicans. Genetics. 2006;174:1737–1744. doi: 10.1534/genetics.106.063271. [DOI] [PMC free article] [PubMed] [Google Scholar]
  83. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R, 1000 Genome Project Data Processing Subgroup The sequence alignment/Map format and SAMtools. Bioinformatics. 2009;25:2078–2079. doi: 10.1093/bioinformatics/btp352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  84. Li H. Aligning sequence reads clone sequence and assembly contigs with BWA-MEM. arXiv. 2013 https://arxiv.org/abs/1303.3997
  85. Libuda DE, Winston F. Amplification of histone genes by circular chromosome formation in Saccharomyces cerevisiae. Nature. 2006;443:1003–1007. doi: 10.1038/nature05205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  86. Lin Z, Li WH. Expansion of hexose transporter genes was associated with the evolution of aerobic fermentation in yeasts. Molecular Biology and Evolution. 2011;28:131–142. doi: 10.1093/molbev/msq184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  87. Lobachev KS, Gordenin DA, Resnick MA. The Mre11 complex is required for repair of hairpin-capped double-strand breaks and prevention of chromosome rearrangements. Cell. 2002;108:183–193. doi: 10.1016/S0092-8674(02)00614-1. [DOI] [PubMed] [Google Scholar]
  88. Lockhart SR, Etienne KA, Vallabhaneni S, Farooqi J, Chowdhary A, Govender NP, Colombo AL, Calvo B, Cuomo CA, Desjardins CA, Berkow EL, Castanheira M, Magobo RE, Jabeen K, Asghar RJ, Meis JF, Jackson B, Chiller T, Litvintseva AP. Simultaneous emergence of Multidrug-Resistant Candida auris on 3 Continents Confirmed by Whole-Genome Sequencing and Epidemiological Analyses. Clinical Infectious Diseases. 2017;64:134–140. doi: 10.1093/cid/ciw691. [DOI] [PMC free article] [PubMed] [Google Scholar]
  89. Lopez V, Barinova N, Onishi M, Pobiega S, Pringle JR, Dubrana K, Marcand S. Cytokinesis breaks dicentric chromosomes preferentially at Pericentromeric regions and telomere fusions. Genes & Development. 2015;29:322–336. doi: 10.1101/gad.254664.114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  90. Lynch M, Sung W, Morris K, Coffey N, Landry CR, Dopman EB, Dickinson WJ, Okamoto K, Kulkarni S, Hartl DL, Thomas WK. A genome-wide view of the spectrum of spontaneous mutations in yeast. PNAS. 2008;105:9272–9277. doi: 10.1073/pnas.0803466105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  91. Maciejowski J, Li Y, Bosco N, Campbell PJ, de Lange T. Chromothripsis and Kataegis induced by telomere crisis. Cell. 2015;163:1641–1654. doi: 10.1016/j.cell.2015.11.054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  92. Magee BB, Magee PT. Induction of mating in Candida Albicans by construction of MTLa and MTLalpha strains. Science. 2000;289:310–313. doi: 10.1126/science.289.5477.310. [DOI] [PubMed] [Google Scholar]
  93. Marotta M, Chen X, Inoshita A, Stephens R, Budd GT, Crowe JP, Lyons J, Kondratova A, Tubbs R, Tanaka H. A common copy-number breakpoint of ERBB2 amplification in breast Cancer colocalizes with a complex block of segmental duplications. Breast Cancer Research. 2012;14:R150. doi: 10.1186/bcr3362. [DOI] [PMC free article] [PubMed] [Google Scholar]
  94. Marotta M, Onodera T, Johnson J, Budd GT, Watanabe T, Cui X, Giuliano AE, Niida A, Tanaka H. Palindromic amplification of the ERBB2 oncogene in primary HER2-positive breast tumors. Scientific Reports. 2017;7:41921. doi: 10.1038/srep41921. [DOI] [PMC free article] [PubMed] [Google Scholar]
  95. Mayer FL, Wilson D, Hube B. Hsp21 potentiates antifungal drug tolerance in candida albicans. PLOS ONE. 2013;8:e60417. doi: 10.1371/journal.pone.0060417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  96. McCoy KM, Tubman ES, Claas A, Tank D, Clancy SA, O'Toole ET, Berman J, Odde DJ. Physical limits on kinesin-5-mediated chromosome congression in the smallest mitotic spindles. Molecular Biology of the Cell. 2015;26:3999–4014. doi: 10.1091/mbc.E14-10-1454. [DOI] [PMC free article] [PubMed] [Google Scholar]
  97. McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, Garimella K, Altshuler D, Gabriel S, Daly M, DePristo MA. The genome analysis toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Research. 2010;20:1297–1303. doi: 10.1101/gr.107524.110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  98. Melnyk AH, Wong A, Kassen R. The fitness costs of antibiotic resistance mutations. Evolutionary Applications. 2015;8:273–283. doi: 10.1111/eva.12196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  99. Mizuno K, Miyabe I, Schalbetter SA, Carr AM, Murray JM. Recombination-restarted replication makes inverted chromosome fusions at inverted repeats. Nature. 2013;493:246–249. doi: 10.1038/nature11676. [DOI] [PMC free article] [PubMed] [Google Scholar]
  100. Møller HD, Parsons L, Jørgensen TS, Botstein D, Regenberg B. Extrachromosomal circular DNA is common in yeast. PNAS. 2015;112:E3114–E3122. doi: 10.1073/pnas.1508825112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  101. Møller HD, Mohiyuddin M, Prada-Luengo I, Sailani MR, Halling JF, Plomgaard P, Maretty L, Hansen AJ, Snyder MP, Pilegaard H, Lam HYK, Regenberg B. Circular DNA elements of chromosomal origin are common in healthy human somatic tissue. Nature Communications. 2018;9:1069. doi: 10.1038/s41467-018-03369-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  102. Morio F, Loge C, Besse B, Hennequin C, Le Pape P. Screening for amino acid substitutions in the candida albicans Erg11 protein of azole-susceptible and azole-resistant clinical isolates: new substitutions and a review of the literature. Diagnostic Microbiology and Infectious Disease. 2010;66:373–384. doi: 10.1016/j.diagmicrobio.2009.11.006. [DOI] [PubMed] [Google Scholar]
  103. Morschhäuser J, Barker KS, Liu TT, BlaB-Warmuth J, Homayouni R, Rogers PD. The transcription factor Mrr1p controls expression of the MDR1 efflux pump and mediates multidrug resistance in candida albicans. PLOS Pathogens. 2007;3:e164. doi: 10.1371/journal.ppat.0030164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  104. Mount HO, Revie NM, Todd RT, Anstett K, Collins C, Costanzo M, Boone C, Robbins N, Selmecki A, Cowen LE. Global analysis of genetic circuitry and adaptive mechanisms enabling resistance to the azole antifungal drugs. PLOS Genetics. 2018;14:e1007319. doi: 10.1371/journal.pgen.1007319. [DOI] [PMC free article] [PubMed] [Google Scholar]
  105. Muñoz JF, Gade L, Chow NA, Loparev VN, Juieng P, Berkow EL, Farrer RA, Litvintseva AP, Cuomo CA. Genomic insights into multidrug-resistance, mating and virulence in Candida Auris and related emerging species. Nature Communications. 2018;9:5346. doi: 10.1038/s41467-018-07779-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  106. Narayanan V, Mieczkowski PA, Kim HM, Petes TD, Lobachev KS. The pattern of gene amplification is determined by the chromosomal location of hairpin-capped breaks. Cell. 2006;125:1283–1296. doi: 10.1016/j.cell.2006.04.042. [DOI] [PubMed] [Google Scholar]
  107. Nattestad M, Alford MC, Sedlazeck FJ, Schatz MC. SplitThreader: exploration and analysis of rearrangements in Cancer genomes. bioRxiv. 2016 doi: 10.1101/087981. [DOI]
  108. Ngamskulrungroj P, Chang Y, Hansen B, Bugge C, Fischer E, Kwon-Chung KJ. Characterization of the chromosome 4 genes that affect fluconazole-induced disomy formation in cryptococcus neoformans. PLOS ONE. 2012;7:e33022. doi: 10.1371/journal.pone.0033022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  109. Niimi K, Monk BC, Hirai A, Hatakenaka K, Umeyama T, Lamping E, Maki K, Tanabe K, Kamimura T, Ikeda F, Uehara Y, Kano R, Hasegawa A, Cannon RD, Niimi M. Clinically significant micafungin resistance in Candida Albicans involves modification of a glucan synthase catalytic subunit GSC1 (FKS1) allele followed by loss of heterozygosity. Journal of Antimicrobial Chemotherapy. 2010;65:842–852. doi: 10.1093/jac/dkq073. [DOI] [PubMed] [Google Scholar]
  110. Nobile CJ, Bruno VM, Richard ML, Davis DA, Mitchell AP. Genetic control of chlamydospore formation in candida albicans. Microbiology. 2003;149:3629–3637. doi: 10.1099/mic.0.26640-0. [DOI] [PubMed] [Google Scholar]
  111. Nobile CJ, Fox EP, Nett JE, Sorrells TR, Mitrovich QM, Hernday AD, Tuch BB, Andes DR, Johnson AD. A recently evolved transcriptional network controls biofilm development in candida albicans. Cell. 2012;148:126–138. doi: 10.1016/j.cell.2011.10.048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  112. Notta F, Chan-Seng-Yue M, Lemire M, Li Y, Wilson GW, Connor AA, Denroche RE, Liang SB, Brown AM, Kim JC, Wang T, Simpson JT, Beck T, Borgida A, Buchner N, Chadwick D, Hafezi-Bakhtiari S, Dick JE, Heisler L, Hollingsworth MA, Ibrahimov E, Jang GH, Johns J, Jorgensen LG, Law C, Ludkovski O, Lungu I, Ng K, Pasternack D, Petersen GM, Shlush LI, Timms L, Tsao MS, Wilson JM, Yung CK, Zogopoulos G, Bartlett JM, Alexandrov LB, Real FX, Cleary SP, Roehrl MH, McPherson JD, Stein LD, Hudson TJ, Campbell PJ, Gallinger S. A renewed model of pancreatic Cancer evolution based on genomic rearrangement patterns. Nature. 2016;538:378–382. doi: 10.1038/nature19823. [DOI] [PMC free article] [PubMed] [Google Scholar]
  113. O'Meara TR, Robbins N, Cowen LE. The Hsp90 chaperone network modulates candida virulence traits. Trends in Microbiology. 2017;25:809–819. doi: 10.1016/j.tim.2017.05.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  114. Onyewu C, Wormley FL, Perfect JR, Heitman J. The calcineurin target, Crz1, functions in azole tolerance but is not required for virulence of candida albicans. Infection and Immunity. 2004;72:7330–7333. doi: 10.1128/IAI.72.12.7330-7333.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  115. Otto SP. The evolutionary consequences of polyploidy. Cell. 2007;131:452–462. doi: 10.1016/j.cell.2007.10.022. [DOI] [PubMed] [Google Scholar]
  116. Paulsen T, Kumar P, Koseoglu MM, Dutta A. Discoveries of extrachromosomal circles of DNA in normal and tumor cells. Trends in Genetics. 2018;34:270–278. doi: 10.1016/j.tig.2017.12.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  117. Pavelka N, Rancati G, Zhu J, Bradford WD, Saraf A, Florens L, Sanderson BW, Hattem GL, Li R. Aneuploidy confers quantitative proteome changes and phenotypic variation in budding yeast. Nature. 2010;468:321–325. doi: 10.1038/nature09529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  118. Payen C, Di Rienzi SC, Ong GT, Pogachar JL, Sanchez JC, Sunshine AB, Raghuraman MK, Brewer BJ, Dunham MJ. The dynamics of diverse segmental amplifications in populations of Saccharomyces cerevisiae adapting to strong selection. G3: Genes, Genomes, Genetics. 2014;4:399–409. doi: 10.1534/g3.113.009365. [DOI] [PMC free article] [PubMed] [Google Scholar]
  119. Perea S, Patterson TF. Antifungal resistance in pathogenic fungi. Clinical Infectious Diseases. 2002;35:1073–1080. doi: 10.1086/344058. [DOI] [PubMed] [Google Scholar]
  120. Perlin DS. Current perspectives on echinocandin class drugs. Future Microbiology. 2011;6:441–457. doi: 10.2217/fmb.11.19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  121. Pfaller MA, Castanheira M, Messer SA, Moet GJ, Jones RN. Variation in candida spp distribution and antifungal resistance rates among bloodstream infection isolates by patient age: report from the SENTRY antimicrobial surveillance program (2008-2009) Diagnostic Microbiology and Infectious Disease. 2010;68:278–283. doi: 10.1016/j.diagmicrobio.2010.06.015. [DOI] [PubMed] [Google Scholar]
  122. Pfaller MA. Antifungal drug resistance mechanisms, epidemiology, and consequences for treatment. The American Journal of Medicine. 2012;125:S3–S13. doi: 10.1016/j.amjmed.2011.11.001. [DOI] [PubMed] [Google Scholar]
  123. Pfaller MA, Diekema DJ, Turnidge JD, Castanheira M, Jones RN. Twenty years of the SENTRY antifungal surveillance program results for Candida Species From 1997-2016. Open Forum Infectious Diseases. 2019;6:S79–S94. doi: 10.1093/ofid/ofy358. [DOI] [PMC free article] [PubMed] [Google Scholar]
  124. Poláková S, Blume C, Zárate JA, Mentel M, Jørck-Ramberg D, Stenderup J, Piskur J. Formation of new chromosomes as a virulence mechanism in yeast candida glabrata. PNAS. 2009;106:2688–2693. doi: 10.1073/pnas.0809793106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  125. Putnam CD, Pallis K, Hayes TK, Kolodner RD. DNA repair pathway selection caused by defects in TEL1, SAE2, and de novo telomere addition generates specific chromosomal rearrangement signatures. PLOS Genetics. 2014;10:e1004277. doi: 10.1371/journal.pgen.1004277. [DOI] [PMC free article] [PubMed] [Google Scholar]
  126. Qiao G, Liu M, Song K, Li H, Yang H, Yin Y, Zhuo R. Phenotypic and comparative transcriptome analysis of different ploidy plants in Dendrocalamus latiflorus Munro. Frontiers in Plant Science. 2017;8:1371. doi: 10.3389/fpls.2017.01371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  127. Ramakrishnan S, Kockler Z, Evans R, Downing BD, Malkova A. Single-strand annealing between inverted DNA repeats: pathway choice, participating proteins, and genome destabilizing consequences. PLOS Genetics. 2018;14:e1007543. doi: 10.1371/journal.pgen.1007543. [DOI] [PMC free article] [PubMed] [Google Scholar]
  128. Ramocki MB, Peters SU, Tavyev YJ, Zhang F, Carvalho CM, Schaaf CP, Richman R, Fang P, Glaze DG, Lupski JR, Zoghbi HY. Autism and other neuropsychiatric symptoms are prevalent in individuals with MeCP2 duplication syndrome. Annals of Neurology. 2009;66:771–782. doi: 10.1002/ana.21715. [DOI] [PMC free article] [PubMed] [Google Scholar]
  129. Rattray AJ, Shafer BK, Neelam B, Strathern JN. A mechanism of palindromic gene amplification in Saccharomyces cerevisiae. Genes & Development. 2005;19:1390–1399. doi: 10.1101/gad.1315805. [DOI] [PMC free article] [PubMed] [Google Scholar]
  130. Reedy JL, Floyd AM, Heitman J. Mechanistic plasticity of sexual reproduction and meiosis in the candida pathogenic species complex. Current Biology. 2009;19:891–899. doi: 10.1016/j.cub.2009.04.058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  131. Riehle MM, Bennett AF, Long AD. Genetic architecture of thermal adaptation in Escherichia coli. PNAS. 2001;98:525–530. doi: 10.1073/pnas.98.2.525. [DOI] [PMC free article] [PubMed] [Google Scholar]
  132. Rodić N, Burns KH. Long interspersed element-1 (LINE-1): passenger or driver in human neoplasms? PLOS Genetics. 2013;9:e1003402. doi: 10.1371/journal.pgen.1003402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  133. Roemer T, Krysan DJ. Antifungal drug development: challenges, unmet clinical needs, and new approaches. Cold Spring Harbor Perspectives in Medicine. 2014;4:a019703. doi: 10.1101/cshperspect.a019703. [DOI] [PMC free article] [PubMed] [Google Scholar]
  134. Ropars J, Maufrais C, Diogo D, Marcet-Houben M, Perin A, Sertour N, Mosca K, Permal E, Laval G, Bouchier C, Ma L, Schwartz K, Voelz K, May RC, Poulain J, Battail C, Wincker P, Borman AM, Chowdhary A, Fan S, Kim SH, Le Pape P, Romeo O, Shin JH, Gabaldon T, Sherlock G, Bougnoux ME, d'Enfert C. Gene flow contributes to diversification of the major fungal pathogen candida albicans. Nature Communications. 2018;9:2253. doi: 10.1038/s41467-018-04787-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  135. Rosenberg A, Ene IV, Bibi M, Zakin S, Segal ES, Ziv N, Dahan AM, Colombo AL, Bennett RJ, Berman J. Antifungal tolerance is a subpopulation effect distinct from resistance and is associated with persistent candidemia. Nature Communications. 2018;9:2470. doi: 10.1038/s41467-018-04926-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  136. Roth JR, Andersson DI. Adaptive mutation: how growth under selection stimulates lac(+) reversion by increasing target copy number. Journal of Bacteriology. 2004;186:4855–4860. doi: 10.1128/JB.186.15.4855-4860.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  137. Rueda C, Puig-Asensio M, Guinea J, Almirante B, Cuenca-Estrella M, Zaragoza O, Padilla B, Muñoz P, Guinea J, Paño Pardo JR, García-Rodríguez J, García Cerrada C, Fortún J, Martín P, Gómez E, Ryan P, Campelo C, de los Santos Gil I, Buendía V, Gorricho BP, Alonso M, Sanz FS, Aguado JM, Merino P, González Romo F, Gorgolas M, Gadea I, Losa JE, Delgado-Iribarren A, Ramos A, Romero Y, Sánchez Romero I, Zaragoza O, Cuenca-Estrella M, Rodriguez-Baño J, Isabel Suarez A, Loza A, Aller García AI, Martín-Mazuelos E, Pérez de Pipaón MR, Garnacho J, Ortiz C, Chávez M, Maroto FL, Salavert M, Pemán J, Blanquer J, Navarro D, Camarena JJ, Zaragoza R, Abril V, Gimeno C, Hernáez S, Ezpeleta G, Bereciartua E, Hernández Almaraz JL, Montejo M, Rivas RA, Ayarza R, Planes AM, Camps IR, Almirante B, Mensa J, Almela M, Gurgui M, Sánchez-Reus F, Martinez-Montauti J, Sierra M, Horcajada JP, Sorli L, Gómez J, Gené A, Urrea M, Valerio M, Díaz-Martín A, Puchades F, Mularoni A. Evaluation of the possible influence of trailing and paradoxical effects on the clinical outcome of patients with candidemia. Clinical Microbiology and Infection. 2017;23:e41–e49. doi: 10.1016/j.cmi.2016.09.016. [DOI] [PubMed] [Google Scholar]
  138. Rustchenko-Bulgac EP. Variations of Candida Albicans electrophoretic karyotypes. Journal of Bacteriology. 1991;173:6586–6596. doi: 10.1128/JB.173.20.6586-6596.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  139. San Millan A, Escudero JA, Gifford DR, Mazel D, MacLean RC. Multicopy plasmids potentiate the evolution of antibiotic resistance in Bacteria. Nature Ecology & Evolution. 2017;1:10. doi: 10.1038/s41559-016-0010. [DOI] [PubMed] [Google Scholar]
  140. Sanglard D, Ischer F, Marchetti O, Entenza J, Bille J. Calcineurin A of Candida Albicans: involvement in antifungal tolerance, cell morphogenesis and virulence. Molecular Microbiology. 2003;48:959–976. doi: 10.1046/j.1365-2958.2003.03495.x. [DOI] [PubMed] [Google Scholar]
  141. Sato H, Masuda F, Takayama Y, Takahashi K, Saitoh S. Epigenetic inactivation and subsequent heterochromatinization of a centromere stabilize dicentric chromosomes. Current Biology. 2012;22:658–667. doi: 10.1016/j.cub.2012.02.062. [DOI] [PubMed] [Google Scholar]
  142. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A. Fiji: an open-source platform for biological-image analysis. Nature Methods. 2012;9:676–682. doi: 10.1038/nmeth.2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  143. Schubert S, Rogers PD, Morschhäuser J. Gain-of-Function mutations in the transcription factor MRR1 are responsible for overexpression of the MDR1 efflux pump in Fluconazole-Resistant candida dubliniensis strains. Antimicrobial Agents and Chemotherapy. 2008;52:4274–4280. doi: 10.1128/AAC.00740-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  144. Sedlazeck FJ, Rescheneder P, Smolka M, Fang H, Nattestad M, von Haeseler A, Schatz MC. Accurate detection of complex structural variations using single-molecule sequencing. Nature Methods. 2018;15:461–468. doi: 10.1038/s41592-018-0001-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  145. Selmecki A, Bergmann S, Berman J. Comparative genome hybridization reveals widespread aneuploidy in Candida Albicans laboratory strains. Molecular Microbiology. 2005;55:1553–1565. doi: 10.1111/j.1365-2958.2005.04492.x. [DOI] [PubMed] [Google Scholar]
  146. Selmecki A, Forche A, Berman J. Aneuploidy and isochromosome formation in drug-resistant candida albicans. Science. 2006;313:367–370. doi: 10.1126/science.1128242. [DOI] [PMC free article] [PubMed] [Google Scholar]
  147. Selmecki A, Gerami-Nejad M, Paulson C, Forche A, Berman J. An isochromosome confers drug resistance in vivo by amplification of two genes, ERG11 and TAC1. Molecular Microbiology. 2008;68:624–641. doi: 10.1111/j.1365-2958.2008.06176.x. [DOI] [PubMed] [Google Scholar]
  148. Selmecki AM, Dulmage K, Cowen LE, Anderson JB, Berman J. Acquisition of aneuploidy provides increased fitness during the evolution of antifungal drug resistance. PLOS Genetics. 2009;5:e1000705. doi: 10.1371/journal.pgen.1000705. [DOI] [PMC free article] [PubMed] [Google Scholar]
  149. Selmecki A, Forche A, Berman J. Genomic plasticity of the human fungal pathogen candida albicans. Eukaryotic Cell. 2010;9:991–1008. doi: 10.1128/EC.00060-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  150. Selmecki AM, Maruvka YE, Richmond PA, Guillet M, Shoresh N, Sorenson AL, De S, Kishony R, Michor F, Dowell R, Pellman D. Polyploidy can drive rapid adaptation in yeast. Nature. 2015;519:349–352. doi: 10.1038/nature14187. [DOI] [PMC free article] [PubMed] [Google Scholar]
  151. Sezgin M, Sankur B. Survey over image thresholding techniques and quantitative performance evaluation. Journal of Electronic Imaging. 2004;13:146–166. doi: 10.1117/1.1631315. [DOI] [Google Scholar]
  152. Shin JH, Chae MJ, Song JW, Jung SI, Cho D, Kee SJ, Kim SH, Shin MG, Suh SP, Ryang DW. Changes in Karyotype and azole susceptibility of sequential bloodstream isolates from patients with candida glabrata candidemia. Journal of Clinical Microbiology. 2007;45:2385–2391. doi: 10.1128/JCM.00381-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  153. Shlien A, Malkin D. Copy number variations and Cancer. Genome Medicine. 2009;1:62. doi: 10.1186/gm62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  154. Singh RP, Prasad HK, Sinha I, Agarwal N, Natarajan K. Cap2-HAP complex is a critical transcriptional regulator that has dual but contrasting roles in regulation of iron homeostasis in candida albicans. Journal of Biological Chemistry. 2011;286:25154–25170. doi: 10.1074/jbc.M111.233569. [DOI] [PMC free article] [PubMed] [Google Scholar]
  155. Singh B, Wu PJ. Linking the organization of DNA replication with genome maintenance. Current Genetics. 2019;65:677–683. doi: 10.1007/s00294-018-0923-8. [DOI] [PubMed] [Google Scholar]
  156. Sionov E, Chang YC, Garraffo HM, Kwon-Chung KJ. Heteroresistance to fluconazole in cryptococcus neoformans is intrinsic and associated with virulence. Antimicrobial Agents and Chemotherapy. 2009;53:2804–2815. doi: 10.1128/AAC.00295-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  157. Sionov E, Lee H, Chang YC, Kwon-Chung KJ. Cryptococcus neoformans overcomes stress of azole drugs by formation of disomy in specific multiple chromosomes. PLOS Pathogens. 2010;6:e1000848. doi: 10.1371/journal.ppat.1000848. [DOI] [PMC free article] [PubMed] [Google Scholar]
  158. Sionov E, Chang YC, Kwon-Chung KJ. Azole heteroresistance in cryptococcus neoformans: emergence of resistant clones with chromosomal disomy in the mouse brain during fluconazole treatment. Antimicrobial Agents and Chemotherapy. 2013;57:5127–5130. doi: 10.1128/AAC.00694-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  159. Sprouffske K, Wagner A. Growthcurver: an R package for obtaining interpretable metrics from microbial growth curves. BMC Bioinformatics. 2016;17:172. doi: 10.1186/s12859-016-1016-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  160. Stimpson KM, Matheny JE, Sullivan BA. Dicentric chromosomes: unique models to study centromere function and inactivation. Chromosome Research. 2012;20:595–605. doi: 10.1007/s10577-012-9302-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  161. Sullivan BA, Schwartz S. Identification of centromeric antigens in dicentric robertsonian translocations: cenp-c and CENP-E are necessary components of functional centromeres. Human Molecular Genetics. 1995;4:2189–2197. doi: 10.1093/hmg/4.12.2189. [DOI] [PubMed] [Google Scholar]
  162. Sun S, Berg OG, Roth JR, Andersson DI. Contribution of gene amplification to evolution of increased antibiotic resistance in Salmonella typhimurium. Genetics. 2009;182:1183–1195. doi: 10.1534/genetics.109.103028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  163. Suzuki T, Nishibayashi S, Kuroiwa T, Kanbe T, Tanaka K. Variance of ploidy in Candida Albicans. Journal of Bacteriology. 1982;152:893–896. doi: 10.1128/jb.152.2.893-896.1982. [DOI] [PMC free article] [PubMed] [Google Scholar]
  164. Taff HT, Mitchell KF, Edward JA, Andes DR. Mechanisms of candida biofilm drug resistance. Future Microbiology. 2013;8:1325–1337. doi: 10.2217/fmb.13.101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  165. Tanaka H, Cao Y, Bergstrom DA, Kooperberg C, Tapscott SJ, Yao MC. Intrastrand annealing leads to the formation of a large DNA palindrome and determines the boundaries of genomic amplification in human Cancer. Molecular and Cellular Biology. 2007;27:1993–2002. doi: 10.1128/MCB.01313-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  166. Tang YC, Amon A. Gene copy-number alterations: a cost-benefit analysis. Cell. 2013;152:394–405. doi: 10.1016/j.cell.2012.11.043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  167. Thorvaldsdóttir H, Robinson JT, Mesirov JP. Integrative genomics viewer (IGV): high-performance genomics data visualization and exploration. Briefings in Bioinformatics. 2013;14:178–192. doi: 10.1093/bib/bbs017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  168. Todd RT, Forche A, Selmecki A. Ploidy variation in fungi: polyploidy, aneuploidy, and genome evolution. Microbiology Spectrum. 2017;5:2016. doi: 10.1128/microbiolspec.FUNK-0051-2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  169. Todd RT, Wikoff TD, Forche A, Selmecki A. Genome plasticity in Candida Albicans is driven by long repeat sequences. eLife. 2019;8:e45954. doi: 10.7554/eLife.45954. [DOI] [PMC free article] [PubMed] [Google Scholar]
  170. Tuch BB, Galgoczy DJ, Hernday AD, Li H, Johnson AD. The evolution of combinatorial gene regulation in fungi. PLOS Biology. 2008;6:e38. doi: 10.1371/journal.pbio.0060038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  171. Tzung KW, Williams RM, Scherer S, Federspiel N, Jones T, Hansen N, Bivolarevic V, Huizar L, Komp C, Surzycki R, Tamse R, Davis RW, Agabian N. Genomic evidence for a complete sexual cycle in candida albicans. PNAS. 2001;98:3249–3253. doi: 10.1073/pnas.061628798. [DOI] [PMC free article] [PubMed] [Google Scholar]
  172. Vandeputte P, Ferrari S, Coste AT. Antifungal resistance and new strategies to control fungal infections. International Journal of Microbiology. 2012;2012:1–26. doi: 10.1155/2012/713687. [DOI] [PMC free article] [PubMed] [Google Scholar]
  173. Venkataram S, Dunn B, Li Y, Agarwala A, Chang J, Ebel ER, Geiler-Samerotte K, Hérissant L, Blundell JR, Levy SF, Fisher DS, Sherlock G, Petrov DA. Development of a comprehensive Genotype-to-Fitness map of Adaptation-Driving mutations in yeast. Cell. 2016;166:1585–1596. doi: 10.1016/j.cell.2016.08.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  174. White TC. The presence of an R467K amino acid substitution and loss of allelic variation correlate with an azole-resistant lanosterol 14alpha demethylase in Candida Albicans. Antimicrobial Agents and Chemotherapy. 1997;41:1488–1494. doi: 10.1128/AAC.41.7.1488. [DOI] [PMC free article] [PubMed] [Google Scholar]
  175. Wu S, Turner KM, Nguyen N, Raviram R, Erb M, Santini J, Luebeck J, Rajkumar U, Diao Y, Li B, Zhang W, Jameson N, Corces MR, Granja JM, Chen X, Coruh C, Abnousi A, Houston J, Ye Z, Hu R, Yu M, Kim H, Law JA, Verhaak RGW, Hu M, Furnari FB, Chang HY, Ren B, Bafna V, Mischel PS. Circular ecDNA promotes accessible chromatin and high oncogene expression. Nature. 2019;575:699–703. doi: 10.1038/s41586-019-1763-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  176. Yang F, Teoh F, Tan ASM, Cao Y, Pavelka N, Berman J. Aneuploidy enables Cross-Adaptation to unrelated drugs. Molecular Biology and Evolution. 2019;36:1768–1782. doi: 10.1093/molbev/msz104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  177. Yona AH, Manor YS, Herbst RH, Romano GH, Mitchell A, Kupiec M, Pilpel Y, Dahan O. Chromosomal duplication is a transient evolutionary solution to stress. PNAS. 2012;109:21010–21015. doi: 10.1073/pnas.1211150109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  178. Zack TI, Schumacher SE, Carter SL, Cherniack AD, Saksena G, Tabak B, Lawrence MS, Zhsng CZ, Wala J, Mermel CH, Sougnez C, Gabriel SB, Hernandez B, Shen H, Laird PW, Getz G, Meyerson M, Beroukhim R. Pan-cancer patterns of somatic copy number alteration. Nature Genetics. 2013;45:1134–1140. doi: 10.1038/ng.2760. [DOI] [PMC free article] [PubMed] [Google Scholar]
  179. Zarrei M, MacDonald JR, Merico D, Scherer SW. A copy number variation map of the human genome. Nature Reviews Genetics. 2015;16:172–183. doi: 10.1038/nrg3871. [DOI] [PubMed] [Google Scholar]
  180. Zeitlin SG, Baker NM, Chapados BR, Soutoglou E, Wang JY, Berns MW, Cleveland DW. Double-strand DNA breaks recruit the centromeric histone CENP-A. PNAS. 2009;106:15762–15767. doi: 10.1073/pnas.0908233106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  181. Zhao Y, Strope PK, Kozmin SG, McCusker JH, Dietrich FS, Kokoska RJ, Petes TD. Structures of naturally evolved CUP1 tandem arrays in yeast indicate that these arrays are generated by unequal nonhomologous recombination. G3: Genes, Genomes, Genetics. 2014;4:2259–2269. doi: 10.1534/g3.114.012922. [DOI] [PMC free article] [PubMed] [Google Scholar]
  182. Zhou J, Lemos B, Dopman EB, Hartl DL. Copy-number variation: the balance between gene dosage and expression in Drosophila melanogaster. Genome Biology and Evolution. 2011;3:1014–1024. doi: 10.1093/gbe/evr023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  183. Żmieńko A, Samelak A, Kozłowski P, Figlerowicz M. Copy number polymorphism in plant genomes. Theoretical and Applied Genetics. 2014;127:1–18. doi: 10.1007/s00122-013-2177-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  184. Zolan ME. Chromosome-length polymorphism in fungi. Microbiological Reviews. 1995;59:686–698. doi: 10.1128/MMBR.59.4.686-698.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]

Decision letter

Editor: Kevin J Verstrepen1

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

We appreciate your efforts to address all the comments raised by the three reviewers, and we like how your work further unravels how medium-sized repeats in the C. albicans genome contribute to its capacity to quickly and reversibly adapt to certain stresses by stimulating amplification as well as loss of genomic regions.

Decision letter after peer review:

Thank you for submitting your article "Expandable and reversible copy number amplification drives rapid adaptation to antifungal drugs" for consideration by eLife. Your article has been reviewed by three peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by Patricia Wittkopp as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: David Gresham (Reviewer #2).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

We would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). Specifically, we are asking editors to accept without delay manuscripts, like yours, that they judge can stand as eLife papers without additional data, even if they feel that they would make the manuscript stronger. Thus the revisions requested below only address clarity and presentation.

We appreciate how this new study expands on your earlier work by showing how repeats are instrumental in facilitating transient amplifications of certain genomic loci that can help C. albicans become tolerant to drugs.

As you can see below, all three reviewers agree that your paper is interesting and well written, and we do not see a necessity for additional experiments. That said, all reviewers did mention that some points merit some additional clarification or discussion. Whereas some comments overlap, other points were only raised by one reviewer, likely reflecting the different expertise of the reviewers. However, we agree that all points that are raised are in fact valid and may help you to improve the paper and make it (even) more accessible and interesting to a broad readership. Hence, in this case, it seemed useful to copy all three separate reviews below instead of merging them all into one common file.

Reviewer #1:

In a previous study (eLife https://doi.org/10.7554/eLife.45954.001), this team showed how long repeat sequences scattered throughout the C. albicans genome fuel recombination events that lead to CNV and LOH. In this follow-up study, the authors show how subjecting C. albicans to antifungal drugs (fluconazole) leads to the selection of CNV in the form of large segmental duplications. The amplified regions ranged from 100 to 1000kb, often contain genes that have previously been associated with drug resistance, and some were amplified up to 12 times. Interestingly, the breakpoints always occurred within 2kb of long inverted repeat sequences, again suggesting that these repeats drive the recombination events. Moreover, removal of the antifungal drugs led to loss of the duplicated regions and return to the initial geno- and phenotype, suggesting that the extra copies impose a negative fitness effect the repeats also facilitate the loss of repeats.

Overall, whereas the idea of 1) CNVs as an adaptive process and 2) repeats as facilitators of genome instability are not new, this study is a useful follow up to the previously published work because it provides more mechanistic insight into the recombination events that allow adaptation to antifungals through swift amplification of certain chromosomal regions. Moreover, the study also shows that the amplification confers a fitness defect in environments without drugs, and how the amplification is quickly lost when the selective pressure disappears. Together, this reveals how medium-sized repeats in the C. albicans genome contribute to its capacity to quickly (and reversibly) adapt to certain stresses by stimulating amplification as well as loss of genomic regions.

Reviewer #2:

In this paper, Todd and Selmecki report copy number variants (CNVs) that arise in response to selection for antifungal resistance. The authors performed 48 evolution experiments in four different genetic backgrounds. Seven isolates that showed increased drug resistance were analyzed using DNA sequencing and CHEF gels. They identify high copy number amplifications of chromosomal segments in resistant clones that do not result from previously reported mechanisms (i.e. centromere amplifications, aneuploidy and use of major repeat sequences (MRS)). The novel CNVs have breakpoints that are near long repeat sequences and are characterized by an intrachromosomal symmetric stair-step pattern of copy number increase. The amplified regions contain several genes and the authors identify likely targets of selection including MRR1, which regulates drug efflux pump expression. Detailed analysis of one set of CNVs provides evidence for the formation of a dicentric region and repeat expansion through a breakage-fusion-bridge mechanism. They then show that when the selection is removed, CNVs rapidly revert.

This is an interesting study. Understanding mechanisms of antifungal resistance is of clear significance and this paper adds to our understanding of the diversity of mechanisms that can be used to generate CNVs underlying drug resistance in C. albicans. Although I do not think that the finding is sufficiently novel to warrant publication as a primary research paper in eLife, the report extends the initial publication by the authors on repeat sequences that mediate CNV formation in the C. albicans genome and therefore is consistent with the aims of the Research Advance format at eLife.

1) It is unclear to me the difference between a major repeat sequence (MJR) and a repeat sequence. The authors state that these breakpoints do not occur near MJR, but do occur "within 2 kb of a long repeat sequence". What is the difference?

2) The paper does not make clear the rate/frequency of this type of CNV forming mechanism. 48 evolution experiments were performed, but only seven clones were isolated for analysis. It's not clear if this report is part of a larger study and this is just a subset of the result. That should be made clearer and if it is part of a larger study the frequency of this mechanism, relative to other mechanisms that generated CNVs at these loci, reported.

3) The symmetrical stair-step property is really striking, but in some cases this seems imperfect (e.g. 2-4-6-5-3-2). How can the authors rule out that this isn't due to noise in the read depth analysis and this should actually be 2-4-6-4-2 or 2-3-5-3-2? i.e. what is the error on the copy number estimation? Is the proposed model expected to generate perfect symmetry?

4) It is surprising that the evolved strains do not have a growth rate defect in rich media. Does this suggest that there is not selection for the loss of the CNV, but simply that it occurs at a very high rate?

5) I do not understand the long read sequencing or its interpretation as it is poorly described. How many isolates were analyzed (the Materials and methods section indicates just one)? How can multiple structural variants be detected at a single locus (e.g. eight detected near repeat 127). I think the authors should improve the description and interpretation of these results.

6) The structure identified from long read sequencing sounds like that found for ODIRA breakpoints (reads switching from the complement to the reverse complement). In addition the stair-step structure is seen at some of the ODIRA-mediated CNVs in budding yeast. Can the authors exclude this DNA replication based mechanism as a possible mechanism forming these CNVs?

Reviewer #3:

In this Research Advance article, authors Todd and Selmecki identify and characterize several new types of fluconazole/azole-resistant isolates of C. albicans that evolve rapidly during growth in the presence of the drug(s) in vitro. Several of these isolates display a remarkable, novel mechanism of drug resistance and tolerance that involves amplification of the copy number of certain chromosome segments by a proposed breakage-fusion-bridge (BFB) model involving the repeated formation and repair of dicentric chromosome intermediates by non-allelic homologous recombination. This BFB model is consistent with the evidence presented and has been proposed to explain similar CNV amplification events in human cancers. The copy number amplification observed in this study is dependent on the presence of two sets of inverted long repeat sequences, which supports the Todd et al., 2019 publication in which previously unidentified long repeat sequences were shown to coincide frequently with breakpoints for CNV and LOH recombination events. This manuscript offers a robust, in-depth characterization of a new mechanism for rapid, adaptive (and reversible) genomic change that builds upon and expands our knowledge of the mechanisms contributing to drug resistance and genome plasticity in human fungal pathogens.

1) This study provides direct evidence of in vitro evolution by CNV amplification resulting in azole drug resistance/tolerance in C. albicans. One limitation of the study is that there is no evidence presented of C. albicans isolates evolving azole resistance via this mechanism in the context of a host infection. The authors could offer comments on how this model could be validated by either direct testing of clinical samples from patients treated with fluconazole or in a host model of infection. Also, in the context of a host infection, would there be any limitations to the formation of dicentric intermediates and breakage-fusion-bridge cycles (or would these events be more likely due to host-mediated DNA damaging agents)? Would cells survive and proliferate sufficiently to generate the copy number amplification necessary for resistance by this mechanism before being targeted for clearance by immune cells, particularly in the presence of fluconazole drug treatment that slows growth of the pathogen?

2) "Separation of Chrs 4-7 identified a dramatic increase in Chr4 size in two isolates (AMS4702 and AMS4444) with Chr4L CNVs relative to their progenitors and indicates that these CNVs are intra-chromosomal, rather than extra-chromosomal, amplifications (Figure 1—figure supplement 2)." In Figure 1—figure supplement 2, there is an absence of a band at Chr4 (denoted by the asterisk) compared to the progenitor for both isolates, but no visible "larger band" that would correspond to the noted "dramatic increase in Chr4 size for these isolates." Please explain (e.g. is the size of Chr4 too large to resolve/migrate through the gel?).

3) It is unclear in the text and figures (such as Figure 6) how the asymmetric stair-step copy numbers arise. A figure depicting symmetric vs. asymmetric stair-step copy number amplification would be helpful.

4) At least some work from budding and fission yeast should be cited (Symington, Kolodner, Lobachev), where genetic control has been examined in some cases. The model with inverted repeats usually involves some type of fold back invasion of an inverted repeat on the same chromosome, followed by BIR to the end, to generate a dicentric.

An example from Kolodner: Putnam et al., 2014.

Fold-backs have also been examined in fission yeast, and these initiated in the context of replication restart:

Mizuno et al., 2013

eLife. 2020 Jul 20;9:e58349. doi: 10.7554/eLife.58349.sa2

Author response


Reviewer #2:

[…] This is an interesting study. Understanding mechanisms of antifungal resistance is of clear significance and this paper adds to our understanding of the diversity of mechanisms that can be used to generate CNVs underlying drug resistance in C. albicans. Although I do not think that the finding is sufficiently novel to warrant publication as a primary research paper in eLife, the report extends the initial publication by the authors on repeat sequences that mediate CNV formation in the C. albicans genome and therefore is consistent with the aims of the Research Advance format at eLife.

1) It is unclear to me the difference between a major repeat sequence (MJR) and a repeat sequence. The authors state that these breakpoints do not occur near MJR, but do occur "within 2 kb of a long repeat sequence". What is the difference?

Thank you for indicating that this needed clarification. The Major Repeat Sequence (MRS) in Candida albicans is a well-characterized long track (10-100 kb) of nested DNA repeats, found 8 times throughout the genome on 7 of the 8 chromosomes. Previous work identified that the MRS is involved in karyotype rearrangements (Lephart and Magee, 2006). The repeats identified in Todd et al., 2019 are distinct from the MRS class of repeats, and were associated with CNVs, inversions, and loss of heterozygosity in a diverse collection of isolates. We have changed the following sentences to clarify the difference between the MRS and the long repeat sequences.

“All copy number breakpoints occurred within 2 kb of one of the 1974 long repeat sequences identified previously by Todd et al., 2019 (Supplementary file 2).”

“…and were not associated with any centromere (CEN) sequence or the C. albicans repetitive element known as the Major Repeat Sequence (MRS, found eight times within the C. albicans genome) (Chibana et al., 1994; Chindamporn et al., 1998; Lephart et al., 2006).”

2) The paper does not make clear the rate/frequency of this type of CNV forming mechanism. 48 evolution experiments were performed, but only seven clones were isolated for analysis. It's not clear if this report is part of a larger study and this is just a subset of the result. That should be made clearer and if it is part of a larger study the frequency of this mechanism, relative to other mechanisms that generated CNVs at these loci, reported.

Thank you for this suggestion. We are very interested in the rate/frequency of these events, especially given that genetically diverse clinical isolates can amplify recurrent CNVs (Figure 1) and that the breakpoints are hotspots for additional genome rearrangements like loss of heterozygosity and inversions in animal models (Todd et al., 2019). Because this is the first study to identify these complex CNVs in C. albicans, a robust, in-depth characterization the CNVs and the mechanisms driving their formation is a necessary first step. The seven isolates described are a part of a larger study into the mechanisms that drive antifungal drug resistance in C. albicans, however most of the remaining populations have not been analyzed yet by whole genome sequencing. Of the 48 FLC-evolved populations, only 14 have been sequenced and 7/14 have complex CNVs. We now include this frequency in the Results section (subsection “Extensive copy number amplifications occur during adaptation to antifungal stress”).

Future studies using single cell analyses with selectable markers are needed to determine the rate of CNV formation during antifungal drug selection and the frequency of these events relative to other adaptive mechanisms. We address the limitations of this study and the need for future experiments that directly quantify the rate of complex CNVs during antifungal drug selection in a new paragraph in the Discussion section (subsection “Model for Complex CNV formation”, second paragraph).

3) The symmetrical stair-step property is really striking, but in some cases this seems imperfect (e.g. 2-4-6-5-3-2). How can the authors rule out that this isn't due to noise in the read depth analysis and this should actually be 2-4-6-4-2 or 2-3-5-3-2? i.e. what is the error on the copy number estimation? Is the proposed model expected to generate perfect symmetry?

First, to clarify, we observed both symmetric (e.g. AMS4702) and asymmetric (e.g. AMS4105) stair-step amplifications. All symmetric CNVs were associated with two inverted repeat sequences, while the asymmetric CNVs were associated with three inverted repeat sequences. We propose that the asymmetric stair-step CNVs arise via a similar dicentric mechanism as symmetric CNVs, however instead of the dicentric chromosome breaking centromere-proximal it breaks within the chromosome arm and leads to NAHR between different repeat sequences. We added Figure 6—figure supplement 1 to describe an example of asymmetric stair-step formation.

Second, we agree that whole genome sequencing read depth data inherently includes some noise. To calculate the copy numbers, read depth for every position in the genome was determined using samtools. Average read depth was calculated using a 500 bp sliding window. The copy number was calculated by normalizing the average read depth to the average nuclear genome depth and multiplying by two. The copy number for each stair-step of the complex CNV was calculated by averaging the copy number for all 500 bp windows contained in each stair-step. To determine the error of the copy number of each stair-step, we calculated the standard error of the mean (SEM). The SEMs ranged from 0.005 – 0.383 (SEM was added to Supplementary file 6). To determine if the SEM was consistent across all stair-steps of the complex CNVs, we determined if there was a correlation between copy number and SEM or stair-step length and SEM. There was a positive correlation between copy number and SEM (p < 0.002, r2 = 0.21) and a negative correlation with stair-step length and SEM (p < 0.003, r2 = 0.19), indicating that short, high copy number stair-steps have a higher SEM than longer, lower copy number stair-steps (see example below). To determine if the calculated error would alter copy number, we determined the copy number of each stair-step +/- the SEM, and then normalized to the base diploid genome and then rounded to the nearest integer for biological relevance. Of the 46 copy number stairsteps (Supplementary file 6), only five stair-steps changed copy number upon the incorporation of the SEM. Three of these stair-steps were located between the same long inverted repeat sequence with a short spacer distance (~5 kb) on Chr3R. For example, AMS4104 Chr3R copy numbers +/- SEM were rounded to: 2-7-13-7-2 or 2-8-13-7-2. Importantly, the overall structure of the complex CNV was not altered, only the rounding of this single stair-step within the CNV. Therefore, we are confident in the copy numbers presented in the manuscript that form both symmetric and asymmetric CNVs.

4) It is surprising that the evolved strains do not have a growth rate defect in rich media. Does this suggest that there is not selection for the loss of the CNV, but simply that it occurs at a very high rate?

We agree that this is surprising, given the size (length and amplitude) of the complex CNVs. However, in general, whole chromosome aneuploidies and CNVs that comprise an entire arm of a chromosome (isochromosomes) are well-tolerated in C. albicans (Selmecki et al., 2006; Todd et al., 2019).

Accordingly, the growth rate data support that most of the isolates with a complex CNV do not have a growth rate defect in rich media. Importantly, these complex CNVs are presumed to be maintained on a monocentric chromosome (Figure 1). We do not know the rate in which these CNVs are lost, but there appears to be very little selection for loss based on growth rate. In contrast, isolates with a complex CNV on a dicentric chromosome (and a partial monosomy) have a strong growth defect (Figure 5). Fast growing clones are rapidly isolated from the dicentric isolate (after one plating on rich media) and these clones have lost the dicentric and the partial monosomic chromosomes (via recombination with the remaining full-length homolog). Interestingly, in the absence of drug selection, these fast growing clones include both euploid (WT) clones and clones containing complex CNVs maintained on a monocentric chromosome (like in Figure 1), which further supports that the cells with complex CNVs can compete with WT genotypes.

5) I do not understand the long read sequencing or its interpretation as it is poorly described. How many isolates were analyzed (the Materials and methods section indicates just one)? How can multiple structural variants be detected at a single locus (e.g. eight detected near repeat 127). I think the authors should improve the description and interpretation of these results.

Thank you for these suggestions, as mentioned above in response to reviewer 1: We have made significant changes to Figure 4 and the Figure 4 legend to clarify how the long-read sequencing was used to characterize novel recombination products (fold-back inversions and non-allelic homologous recombination between inverted repeats) in the azole evolved isolate, AMS3051. The only isolate sequenced using the Oxford Nanopore MinION was AMS3051 that contains the Chr3L complex CNV on a dicentric chromosome. We changed the color scheme and added a schematic to show the WT and post-recombination chromosomes. We also modified the Results section to better describe how the long-read sequences supported each of the two structural variants, one at Repeat 124 and one at Repeat 127 (subsection “Recombination occurs between long inverted repeats leading to CNV formation”).

6) The structure identified from long read sequencing sounds like that found for ODIRA breakpoints (reads switching from the complement to the reverse complement). In addition the stair-step structure is seen at some of the ODIRA-mediated CNVs in budding yeast. Can the authors exclude this DNA replication based mechanism as a possible mechanism forming these CNVs?

We agree that these two features (complement/reverse complement switching and stair-step patterns) are similar between the copy number breakpoints described in this manuscript and copy number breakpoints attributed to ODIRA. However, to the best of our knowledge, the other key features of ODIRA are not observed in our data. In the ODIRA model of CNV formation, errors during DNA replication result in the ligation of the leading and lagging strands at short (~8 bp), closely spaced (~40 bp) inverted repeat sequences within the same replication fork. In contrast, the long inverted repeat sequences associated with the complex CNVs in C. albicans are both much longer (median copy length of ~1.7 kb) and separated by a much greater distance (ranging from 1609 bp – 96,231 bp, median spacer distance ~38,000 bp) than the repeats observed in S. cerevisiae. Given the distance between the inverted repeat sequences in C. albicans, it is unlikely that the repeats would be present in the same replication fork and undergo ligation of the leading and lagging strands, as in ODIRA. In addition, we determined the distance between the C. albicans predicted origins of replication (Tsai et al., 2014) and the long repeat sequences associated with the complex CNVs and found that the nearest predicted origins were located between ~600 bp – ~100 kb (median distance ~16 kb) away from the long repeats. Therefore, while we cannot formally exclude a DNA replication based mechanism in the formation of C. albicans complex CNVs, the length and distance between inverted repeats, and distance between the repeats and origins of replication, makes this unlikely.

Reviewer #3:

[…] 1) This study provides direct evidence of in vitro evolution by CNV amplification resulting in azole drug resistance/tolerance in C. albicans. One limitation of the study is that there is no evidence presented of C. albicans isolates evolving azole resistance via this mechanism in the context of a host infection. The authors could offer comments on how this model could be validated by either direct testing of clinical samples from patients treated with fluconazole or in a host model of infection. Also, in the context of a host infection, would there be any limitations to the formation of dicentric intermediates and breakage-fusion-bridge cycles (or would these events be more likely due to host-mediated DNA damaging agents)? Would cells survive and proliferate sufficiently to generate the copy number amplification necessary for resistance by this mechanism before being targeted for clearance by immune cells, particularly in the presence of fluconazole drug treatment that slows growth of the pathogen?

Thank you for suggesting additional discussion of our observations in the context of a host infection. We agree that a limitation of this study is that we do not know the rate of dicentric chromosome and complex CNV formation in vitro or in vivo. in vivo passage of C. albicans through a mouse model of candidiasis (in the absence of antifungal drug) also identified CNVs at long repeat sequences (Todd et al., 2019). This suggests that these long repeat sequences can generate genomic variation in diverse environments, including in vivo. Host factors, such as reactive oxygen species may increase the rate of DSBs that result in dicentric chromosome formation, in the presence or absence of antifungal drugs. Indeed, dicentric chromosomes have been identified in clinical isolates of C. albicans, suggesting that these chromosome structures can form in vivo (Selmecki, 2006). Future studies using single-cell analyses (e.g. Lauer et al., 2018) are needed to determine the rate of complex CNV formation both in vitro and in vivo in the presence and absence of antifungal drugs. We have added a new paragraph to the Discussion section to address these points (subsection “Model for Complex CNV formation”, second paragraph).

2) "Separation of Chrs 4-7 identified a dramatic increase in Chr4 size in two isolates (AMS4702 and AMS4444) with Chr4L CNVs relative to their progenitors and indicates that these CNVs are intra-chromosomal, rather than extra-chromosomal, amplifications (Figure 1—figure supplement 2)." In Figure 1—figure supplement 2, there is an absence of a band at Chr4 (denoted by the asterisk) compared to the progenitor for both isolates, but no visible "larger band" that would correspond to the noted "dramatic increase in Chr4 size for these isolates." Please explain (e.g. is the size of Chr4 too large to resolve/migrate through the gel?).

Thank you for suggesting additional details for this figure. The wild-type Chr4 in C. albicans is ~1.6 MB in length. Given the size of complex CNVs in AMS4702 and AMS4444, the estimated size of Chr4 increases to ~3.4 MB and ~2.8 MB. These larger bands co-migrate with Chr1 and ChrR. To address this, we added the chromosome sizes to the CHEF gel image in Figure 1—figure supplement 2, and updated the figure legend.

3) It is unclear in the text and figures (such as Figure 6) how the asymmetric stair-step copy numbers arise. A figure depicting symmetric vs. asymmetric stair-step copy number amplification would be helpful.

Thank you for this suggestion. Asymmetric stair-steps were associated with three inverted repeat sequences, whereas symmetric stair-steps were associated with only two inverted repeat sequences. We propose that the asymmetric stair-step CNVs arise via a similar dicentric mechanism as symmetric CNVs, however instead of the dicentric chromosome breaking centromere-proximal it breaks within the chromosome arm and leads to NAHR between different repeat sequences. We have added Figure 6—figure supplement 1 as an example of asymmetric stair-step formation.

4) At least some work from budding and fission yeast should be cited (Symington, Kolodner, Lobachev), where genetic control has been examined in some cases. The model with inverted repeats usually involves some type of fold back invasion of an inverted repeat on the same chromosome, followed by BIR to the end, to generate a dicentric.

An example from Kolodner: Putnam et al., 2014.

Fold-backs have also been examined in fission yeast, and these initiated in the context of replication restart: Mizuno et al., 2013.

Thank you for these suggestions, we agree that fold back invasion of a repeat sequence on the same chromosome followed by BIR is as likely as NAHR between repeat sequences on sister chromatids. Because we cannot distinguish between these two events, we now indicate that both are possible in the Discussion (subsection “Model for Complex CNV formation”, first paragraph) and in the Figure 6 legend, and include the above references accordingly. One interesting distinction is that the long inverted repeats described in this study are separated by greater distances (median spacer distance ~38 kb) than repeat sequences typically involved in fold back invasion and BIR in S. cerevisiae and S. pombe.

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Data Citations

    1. Todd RT, Selmecki A. 2020. Complex copy number variants in Candida albicans. NCBI BioProject. PRJNA613282
    2. Mount HO, Revie NM, Todd RT, Anstett K, Collins C, Costanzo M, Boone C, Robbins N, Selmecki A, Cowen LE. 2018. Candida albicans genome sequencing. NCBI BioProject. PRJNA323475 [DOI] [PMC free article] [PubMed]

    Supplementary Materials

    Figure 2—source data 1. Minimum inhibitory concentration raw data.
    Supplementary file 1. Strains used in this study.
    elife-58349-supp1.xlsx (11.1KB, xlsx)
    Supplementary file 2. Features of long-repeat sequences identified at copy number breakpoints of complex CNVs.
    elife-58349-supp2.xlsx (16.5KB, xlsx)
    Supplementary file 3. Growth curve raw data and analysis.
    elife-58349-supp3.xlsx (74.9KB, xlsx)
    Supplementary file 4. Coding sequences within complex CNVs.
    elife-58349-supp4.xlsx (191.3KB, xlsx)
    Supplementary file 5. Primers used in this study.
    elife-58349-supp5.xlsx (32.3KB, xlsx)
    Supplementary file 6. Summary of complex CNV features.
    elife-58349-supp6.xlsx (79.8KB, xlsx)
    Transparent reporting form

    Data Availability Statement

    All genome sequencing data have been deposited in the Sequence Read Archive under BioProject PRJNA613282 All data analyzed during this study are included in the manuscript and supporting files. The source data file is provided for Figure 2.

    The following dataset was generated:

    Todd RT, Selmecki A. 2020. Complex copy number variants in Candida albicans. NCBI BioProject. PRJNA613282

    The following previously published dataset was used:

    Mount HO, Revie NM, Todd RT, Anstett K, Collins C, Costanzo M, Boone C, Robbins N, Selmecki A, Cowen LE. 2018. Candida albicans genome sequencing. NCBI BioProject. PRJNA323475


    Articles from eLife are provided here courtesy of eLife Sciences Publications, Ltd

    RESOURCES