Use of CRISPR-Cas9 To Target Homologous Recombination Limits Transformation-Induced Genomic Changes in Candida albicans

Abstract

Most of our knowledge relating to molecular mechanisms of human fungal pathogenesis in Candida albicans relies on reverse genetics approaches, requiring strain engineering. DNA-mediated transformation of C. albicans has been described as highly mutagenic, potentially accentuated by the organism's genome plasticity, including the acquisition of genomic rearrangements, notably upon exposure to stress. The advent of CRISPR-Cas9 has vastly accelerated the process of genetically modifying strains, especially in diploid (such as C. albicans) and polyploid organisms. The effects of unleashing this nuclease within the genome of C. albicans are unknown, although several studies in other organisms report Cas9-associated toxicity and off-target DNA breaks. Upon the construction of a C. albicans strain collection, we took the opportunity to compare strains which were constructed using CRISPR-Cas9-free and CRISPR-Cas9-dependent transformation strategies, by quantifying and describing transformation-induced loss-of-heterozygosity and hyperploidy events. Our analysis of 57 strains highlights the mutagenic effects of transformation in C. albicans, regardless of the transformation protocol, but also underscores interesting differences in terms of genomic changes between strains obtained using different transformation protocols. Indeed, although strains constructed using the CRISPR-Cas9-free transformation method display numerous concomitant genomic changes randomly distributed throughout their genomes, the use of CRISPR-Cas9 leads to a reduced overall number of genome changes, particularly hyperploidies. Overall, in addition to facilitating strain construction by reducing the number of transformation steps, the CRISPR-Cas9-dependent transformation strategy in C. albicans appears to limit transformation-associated genome changes.IMPORTANCE Genome editing is essential to nearly all research studies aimed at gaining insight into the molecular mechanisms underlying various biological processes, including those in the opportunistic pathogen Candida albicans The adaptation of the CRISPR-Cas9 system greatly facilitates genome engineering in many organisms. However, our understanding of the effects of CRISPR-Cas9 technology on the biology of C. albicans is limited. In this study, we sought to compare the extents of transformation-induced genomic changes within strains engineered using CRISPR-Cas9-free and CRISPR-Cas9-dependent transformation methods. CRISPR-Cas9-dependent transformation allows one to simultaneously target both homologs and, importantly, appears less mutagenic in C. albicans, since strains engineered using CRISPR-Cas9 display an overall decrease in concomitant genomic changes.

Keywords: CRISPR-Cas9; Candida albicans; genome rearrangements.

FIG 1

CRISPR-Cas9-free and -dependent transformation strategies used for strain construction. Derived from the C. albicans reference strain SC5314, the parental SN148 strain was sequentially transformed using CRISPR-Cas9-free (gray) or CRISPR-Cas9-dependent (orange) strategy in order to integrate (i) the BFP/GFP LOH reporter system, (ii) a unique barcode sequence, and (iii) the leucine marker rendering the strains prototrophic. Strains constructed using the CRISPR-Cas9-free strategy underwent four transformation steps, as opposed to two using the CRISPR-Cas9-dependent strategy. The barcode sequence (BC) and LEU2 auxotrophic marker are introduced at the RPS1 locus on Chr1. The BC and LEU2 marker are introduced at another locus for the strains carrying the LOH reporter system on Chr1 [in the intergenic sequence between the CDR3 and tG(GCC)2 loci], as indicated by the striped arrows.

FIG 2

Doubling times in YPD medium at 30°C. Each data point is representative of the average doubling time (in minutes) (n = 8) of the parental (SN148) strain (red square) or a constructed strain (circles, CRISPR-Cas9-free method; diamonds, CRISPR-Cas9-dependent method), with error bars indicating standard deviations. The average doubling time of all constructed strains (57 strains) is represented by the orange horizontal line (with orange shading indicating standard deviations). Strains displaying at least one genomic change are shown in yellow, while black indicates strains that are free of transformation-acquired genomic changes.

FIG 3

Identification of transformation-induced genome changes within the 57 sequenced C. albicans strains. (A) Determining the average ABHet values per chromosome (Chr) for strain CEC5775. (Upper panel) Plots of allele balances for the eight chromosomes of strain CEC5775. (Lower panels) Histograms illustrate the distribution of ABHet values across a given chromosome, where the black vertical bar represents a 0.5 ABHet value (heterozygous diploid). ABHet and ABHom values are shown in blue and red, respectively. Data interpretation for each chromosome is as follows: Chr1, -6, and -7, trisomy (1×HapA, 2×HapB); Chr2 and -3, trisomy (2×HapA, 1×HapB); Chr4, disomy (1×HapA, 1×HapB); Chr5, LOH; ChrR, recombination event localized in proximity of the centromere plus trisomy (left arm, 1×HapA, 2×HapB; right arm, 2×HapA, 1×HapB). Additional LOH events have been described previously in parental strain SN148 (LOH on Chr2) (25) and in SC5314 (LOH on Chr3 and Chr7) (12). (B) Summary of the genomic changes identified across the eight chromosomes for all 57 sequenced C. albicans strains, using the strategy presented in panel A. The plots showing the allele balance at heterozygous positions and the mean ABHet values per chromosome for each strain can be found in Fig. S1 and Table S5, respectively. LOH events are indicated in blue, aneuploidies in orange. Genomic changes impacting whole chromosomes are identified by solid colors, while those partially impacting chromosomes are identified by a striped pattern.

FIG 4

Quantification of genome changes identified within C. albicans strains engineered using one of two transformation methods. (A) Percentages of chromosomes and strains impacted by transformation-associated genomic changes, as well as percentages of genomic-change-free strains. (B) Studying concurrent genomic changes within strains. (C) Percentage of each chromosome affected by at least one genomic change. (D) Representations of the numbers of strains displaying genomic change(s) on a targeted chromosome, a nontargeted chromosome, or both types. The fraction of each category of strains within genomic-change-displaying strains is indicated by a percentage. (E) Frequency of transformation-associated genomic changes tabulated as the average number of chromosomes displaying genome changes per strain per transformation. Differences in genome change frequencies between strains constructed using the two transformation strategies are represented as fold changes (*, P < 0.05; **, P < 0.01; ***, P < 0.001 by t test).

FIG 5

Nature of genomic changes identified within sequenced C. albicans strains. (A) Frequency of transformation-associated genomic changes tabulated as the average number of genome changes per strain per transformation. Differences in genomic change frequencies between the two transformation methods are represented as fold changes (P, ≤0.05 by t test). (B) Percentage of strains displaying each type of genomic change per chromosome and transformation strategy.