Targeted Genome Editing via CRISPR in the Pathogen Cryptococcus neoformans

Low rates of homologous integration have hindered molecular genetic studies in Cryptococcus neoformans over the past 20 years, and new tools that facilitate genome manipulation in this important pathogen are greatly needed. To this end, we have investigated the use of a Class 2 CRISPR system in C. neoformans (formerly C. neoformans var. grubii). We first expressed a derivative of the Streptococcus pyogenes Cas9 nuclease in C. neoformans, and showed that it has no effect on growth, production of virulence factors in vitro, or virulence in a murine inhalation model. In proof of principle experiments, we tested the CAS9 construct in combination with multiple self-cleaving guide RNAs targeting the well-characterized phosphoribosylaminoamidazole carboxylase-encoding ADE2 gene. Utilizing combinations of transient and stable expression of our constructs, we revealed that functionality of our CRISPR constructs in C. neoformans is dependent upon the CAS9 construct being stably integrated into the genome, whilst transient expression of the guide RNA is sufficient to enhance rates of homologous recombination in the CAS9 genetic background. Given that the presence of the CRISPR nuclease does not influence virulence in a murine inhalation model, we have successfully demonstrated that this system is compatible with studies of C. neoformans pathogenesis and represents a powerful tool that can be exploited by researchers in the field.


Introduction
Over the past thirty years the incidence of life-threatening infections caused by opportunistic fungal pathogens has skyrocketed due to the AIDS pandemic and the advent of new immunesuppressing technologies. The majority of these infections are caused by Candida albicans, Aspergillus fumigatus and Cryptococcus neoformans. For those suffering from AIDS, the most dangerous of these pathogens is C. neoformans; the epidemiology of cryptococcosis mirrors that of Human Immunodeficiency Virus, with a formidable incidence rate in countries in which the burden of HIV is high [1,2]. Existing in the environment as a ubiquitous saprophyte, the pathogen disseminates via the lungs to the central nervous system causing life-threatening meningoencephalitis. Even with the best available treatments, mortality in developed countries can be up to 20%. In less privileged nations where treatment is limited, there is a much poorer patient prognosis, with mortality rates as high as 75% [1][2][3][4].
Our understanding of the mechanisms employed by C. neoformans to cause disease in the human host has been greatly advanced through the development of molecular genetic techniques that have facilitated the investigation of virulence determinants. Yet the low rate of homologous recombination during transformation of this organism remains a major hurdle, impeding efficient analysis of gene function [5][6][7][8]. Transformation of C. neoformans was first achieved in 1990 through electroporation [9], however the technique was not widely adopted due to the instability of transformants and low homologous integration rates. Molecular genetic modifications did not become routine in C. neoformans research until after the development of a biolistic transformation protocol in 1993 [8]. While this approach is now utilized extensively in the field, it typically only achieves homologous recombination rates of around 10%.
The low rate of homologous integration in C. neoformans is due largely to the efficiency of Non-Homologous End Joining (NHEJ) driving ectopic integration as the most common fate of transformed DNA. As in other species, removal of a component of the Ku heterodimer that is required for the first step of NHEJ disables this process, resulting in a higher frequency of homologous integration [6]. Unfortunately, kuΔ strains present a number of issues; KU80 expression is upregulated during infection of a human host, and the ku80Δ mutant is less successful in a murine competition model of infection [10,11]. Disrupting NHEJ through the use of kuΔ mutants is therefore an inappropriate choice for many studies, and as such these strains have not seen widespread use within the C. neoformans community.
One alternative to disrupting NHEJ by disabling Ku is to instead enhance the local rate of homologous integration by introducing a double stranded break, thus activating mechanisms that can integrate transforming constructs through homology during the DNA repair process [12]. Over the last three years, an approach that has been successfully employed in numerous organisms is to create targeted double stranded breaks via Clustered Regularly Interspaced Short Palindromic Repeat (CRISPR) technology. CRISPR is an adaptive immunity mechanism in prokaryotes [13]. While there are a number of classes of CRISPR systems, they all generally act in three stages. In the first stage, acquisition, the microbe identifies non-self DNA from bacteriophages and other invading genetic elements with the aid of CRISPR associated (Cas) proteins. These identify Protospacer Adjacent Motifs (PAMs), usually the sequence NGG (where N stands for any base), and copy~24-48 bp upstream of this motif as a "spacer" into a CRISPR locus between two short (~30 bp) palindromic repeats. The sequential addition of new spacers over time generates CRISPR loci that can contain as many as several hundred repeats [13][14][15][16][17]. In the second stage, expression, the CRISPR locus is transcribed, producing pre-crRNA which is cleaved to form crRNAs consisting of a single spacer and an associated repeat. The crRNA then associates with a nuclease; in Class 1 systems this is a multi-subunit complex (Cascade in Escherichia coli), in Class 2 it is a single protein (Cas9 in Streptococcus pyogenes). In most systems this association is dependent on the crRNA first hybridizing with a complementary transactivating CRISPR RNA (tracrRNA) [18]. In the third and final stage, interference, foreign DNA identical to the crRNA spacer is recognized and cleaved by the nuclease.
First discovered in the bacterium E. coli and the archaea Haloferax mediterranei [19,20], CRISPR systems have now been identified in a wide range of prokaryotes. Following success in harnessing the activity of CRISPR in prokaryotes [21], a minimal system was engineered for use in eukaryotes. Analyses of a range of CRISPR nucleases identified the simple Class 2 enzyme of S. pyogenes as effective in human cells following codon optimization and the attachment of an SV40 nuclear localization sequence. Furthermore, experiments with crRNA and tracrRNA revealed these could be fused, enabling targeting of nuclease activity with a guideRNA (gRNA), a single engineered RNA molecule with characteristics of both the crRNA and tracrRNA components.
The utility of the minimal CRISPR-Cas9 system in enabling targeted gene disruption has now been exploited in a diverse range of eukaryotes, including mice and rats [22][23][24][25], plants [26][27][28] and also the fungal species Saccharomyces cerevisiae [29], Schizosaccharomyces pombe [30], A. fumigatus [31] and C. albicans [32]. The successful use of a CRISPR-Cas9 system has also recently been described for Cryptococcus deneoformans [33] (formerly C. neoformans var. neoformans), a less clinically prevalent member of the Cryptococcus genus [34]. Intriguingly, it was reported that expression of Cas9 in C. neoformans "exhibits a remarkably negative effect on virulence", indicating that this important new and widespread molecular tool cannot be employed in the species most commonly associated with life-threatening cryptococcosis.
We have investigated the application of CRISPR in C. neoformans, creating an expression construct for the S. pyogenes Cas9 nuclease and generating a strain with it integrated at the well-characterized Safe Haven region of chromosome one [35]. Contrary to the previous report, we have demonstrated that constitutive expression of Cas9 in C. neoformans has no effect on growth, virulence factor production or the ability to cause disease in a murine inhalation model of infection. Co-transforming the Cas9 strain simultaneously with deletion and self-cleaving gRNA constructs targeting the phosphoribosylaminoamidazole carboxylaseencoding ADE2 gene dramatically increased the rate of homologous recombination, indicating that this system can be successfully employed in C. neoformans as an important tool to facilitate targeted modifications of the genome.

Strains and growth conditions
C. neoformans strains generated in this study are listed in S1 Table and were grown on YPD agar unless stated otherwise. All strains were created in C. neoformans type strain H99 [36]. Plasmid maintenance and cloning was carried out using the E. coli strain Mach1 (Life Technologies, USA). Plasmids generated in this study are listed in S2 Table. Bioinformatics Sequence analyses were performed using BLASTp and MacVector 10.0 (MacVector Inc., USA), with sequence alignments generated using ClustalW v1.4. Sequence traces generated at the Australian Genome Research Facility (Australia) were analyzed using Sequencher 4.7 (Gene Codes Corporation, USA).

Construction of a C. neoformans strain containing a CAS9 expression cassette
The C. neoformans CAS9 expression construct was created by combining the TEF1 (CNAG_06125) C. neoformans promoter (primers UQ3590 and UQ3591), a human codon optimized, SV40 NLS-containing version of the S. pyogenes CAS9 gene [37] (UQ3592 and UQ3593), and TEF1 terminator (UQ3594 and UQ3595) by overlap PCR (UQ3590 and UQ3595). All PCR was performed using Phusion High-Fidelity DNA polymerase (New England Biolabs, USA) using H99 genomic DNA as the template for TEF1 and plasmid p414-TEF1p-Cas9-CYC1 for CAS9 (Plasmid 43802; Addgene, USA) [29]. Primers are listed in S3 Table. The overlap product was cloned as a NotI/KpnI fragment into the Safe Haven hygromycin B resistance vector pSDMA58 (Plasmid 67942; Addgene, USA) [35] and sequenced. The resulting plasmid pSDMA65 was linearized using BaeI and biolistically transformed into type strain H99, with transformants selected on YPD agar supplemented with 200 μg/mL hygromycin B (Life Technologies, USA) [5]. Integration of the plasmid at the Safe Haven site was confirmed via colony PCR [35] and Southern blot using the Safe Haven fragment as a probe (primers UQ2962 and UQ2963), creating strain H99 CAS9 .

Quantitative real-time PCR
Strains H99 and H99 CAS9 were grown in YPD media shaking at 30°C for 16 hr. Cells were collected, and the resulting pellets frozen and lyophilized, with total RNA isolated using TRIzol Reagent (Life Technologies, USA). cDNA was created using the Superscript III First-Strand Synthesis System (Invitrogen, USA) according to the manufacturer's protocol. Quantitative real-time PCR (qRT-PCR) of CAS9 (primers UQ3813 and UQ3814) was performed using SYBR Green Supermix (Applied Biosystems, USA) and an Applied Biosystems 7900HT Fast Real Time PCR System with the following cycling conditions: denaturation at 95°C for 10 min, amplification and quantification for 45 cycles at 95°C for 15 sec and 60°C for 1 min, with melting-curve profiling of 95°C for 2 min, 60°C for 15 sec, and 95°C for 15 sec. Relative gene expression was quantified using SDS 1.3.1 (Applied Biosystems, USA) based on the 2 −ΔΔCT method [38]. The actin-encoding gene ACT1 (primers UQ482 and UQ728) was used as a control for normalization. One-way analysis of variance was performed using the unpaired, twotailed t-test in GraphPad Prism Version 7.0 (GraphPad Software, USA). P-values of <0.05 were considered statistically significant.

Phenotypic and virulence factor assays
Fresh cultures of H99 and H99 CAS9 grown in YPD were washed and diluted to OD 600 1.0, and 10-fold serially diluted prior to spotting. Melanization assays were performed on solid L-DOPA media supplemented with 10 mM asparagine [39]. Phospholipase B production was visualized on Sabouraud dextrose agar with 8% egg yolk [40], protease production on complete YNB with 1% BSA [41] and urease production on Christensen's agar [42]. Capsule growth was induced by growth for 24 hr in RPMI 1640 media (Life Technologies, USA) with 2% glucose and 10% fetal bovine serum (Life Technologies, USA) and stained with India Ink (BD Diagnostics, USA). Cells were imaged with a Leica DM2500 microscope and DFC425C camera (Leica, Germany). At least 5 independent images were taken and the relative capsule diameter of 25 cells from each culture determined as described by Zaragoza et al. [43]. Capsule diameter was measured relative to cellular diameter using the ruler tool in Adobe Photoshop CS6 (Adobe Systems Incorporated, USA). Experiments were performed in biological triplicate and unpaired, two-tailed t-tests were performed in GraphPad Prism Version 7.0 (GraphPad Software, USA) to compare variation between replicates. All assays were performed at both 30 and 37°C.

Murine inhalation model of cryptococcosis
For murine infection assays, 6-week-old female BALB/c mice (Animal Resources Centre, Australia) were infected by nasal inhalation [44]. For each strain (H99 and H99 CAS9 ), 10 mice were inoculated with a 50 μL drop containing 5 × 10 5 C. neoformans cells. A maximum of 5 mice were housed per IVC cage (Tecniplast, USA) with bed-o' cobs 1/8" bedding (The Andersons, USA), crink-L' nest nesting material (The Andersons, USA), and cardboard as environmental enrichment. Mice were provided Rat and Mouse Cubes (Specialty Feeds, Australia) and water ad libitum. Each mouse was examined and weighed twice daily for the duration of the experiment, with affected mice euthanized via CO 2 inhalation once body weight had decreased to 80% of pre-infection weight or they exhibited symptoms consistent with infection. Death after CO 2 inhalation was confirmed by pedal reflex prior to dissection. The brain, lungs, liver, spleen and kidneys were collected, homogenized and plated to determine colony-forming units per gram organ weight. Kaplan-Meier survival curves were plotted using GraphPad Prism 7.0 (GraphPad Software, USA). Significance was analyzed using the log-rank test, while organ burden significance was determined using a one-way ANOVA with Tukey's multiple comparisons test. P values of <0.05 were considered significant.

Ethics statement
This study was carried out in strict accordance with the recommendations in the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes by the National Health and Medical Research Council. The protocol was approved by the Molecular Biosciences Animal Ethics Committee (AEC) of The University of Queensland (AEC approval no. SCMB/439/13/UQ/NHMRC). Infection was performed under methoxyflurane anaesthesia, and all efforts were made to minimize suffering through adherence to the Guidelines to Promote the Wellbeing of Animals Used for Scientific Purposes as put forward by the National Health and Medical Research Council (Australia).

Construction of ADE2 mutant strains
A deletion construct for the ADE2 gene was generated as per Arras et al. [35]. Briefly, the construct was prepared using overlap PCR, employing primers UQ1439 and UQ1442 to join the 994 bp ADE2 5' region (primers UQ1439 and UQ1440), the G418 resistance marker NEO (UQ2808 and UQ2809) and the ADE2 982 bp 3' region (UQ1441 and UQ1442). H99 genomic DNA was used as the template for ADE2, and the plasmid pJAF1 for NEO [46]. C. neoformans transformations were carried out via biolistic particle delivery as previously described [5] onto media containing 100 μg/mL G418 (Sigma, USA). Correct ade2Δ integrants were initially identified via their pink coloration on YPD media and adenine auxotrophy on YNB media. Representative pink transformants were selected for Southern blot analyses to validate correct integration of the ADE2 deletion cassette using the UQ1439 and UQ1442 PCR product as a probe. Only transformations that resulted in 10 or more colonies were included in subsequent analyses due to larger numbers providing a greater level of precision. To determine significance of correct integration rates between conditions, ANOVA followed by Tukey's HSD tests were completed in GraphPad Prism Version 7.0 (GraphPad Software, USA). P-values of <0.05 were considered statistically significant. The absence of the gRNA plasmids following co-transformation with the ade2::NEO construct was verified via Southern blot, using the UQ234 and UQ235 PCR product generated from plasmid pSDMA64 as a probe.

Results
Creating a CAS9 expression construct for C. neoformans The utility of CRISPR as a molecular tool in eukaryotes was first proven in human cell lines, with two groups developing constructs to express the Class 2 Type II Cas9 from S. pyogenes ( Fig 1A) [37,47]. To begin our studies in C. neoformans, we investigated these humanized CAS9 constructs as a starting point. Comparing the CAS9 coding regions of the humanized constructs revealed only 81.1% nucleotide identity, with over half of the codons differing by at least one nucleotide; neither publication described the method of codon optimization for humans, but they clearly used very different criteria. Of the two, the construct from Mali et al. [37] had both codon usage and GC content more closely aligned to that published for C. deneoformans [48]. These humanized CAS9 constructs also differ in their complexity. Both constructs contain the monopartite SV40 large T-antigen NLS for nuclear localization, but the Cong et al. [47] construct also includes a triplicate FLAG tag, and the bipartite Xenopus laevis nucleoplasmin NLS (Fig 1A). As we could not predict the consequences in C. neoformans of introducing each additional construct element, we considerred the simplicity of the Mali et al. construct to be more suitable for our intended purpose. Furthermore, the Mali et al. construct had been successfully employed in S. cerevisiae, demonstrating it can function in fungi [29].
To adapt the Mali et al. construct for use in C. neoformans, we first selected a promoter and terminator that would enable expression in this pathogen. Based on the success of DiCarlo et al. who used the elongation factor 1-alpha (TEF1) promoter to express CAS9 in S. cerevisiae [29], we opted to use the promoter of the C. neoformans TEF1 homolog. The C. neoformans TEF1 5' UTR begins 500 bp upstream of the ATG and includes an intron. In order to include this long UTR and associated regulatory elements, we PCR amplified the promoter as a 1,256 bp fragment, combined it via overlap PCR with the CAS9-SV40 ORF and the TEF1 terminator region, and subcloned this product into the pSDMA58 Safe Haven vector for targeting to chromosome 1 (Fig 1A) [35]. The construct was then integrated into the genome of C. neoformans type strain H99 to create strain H99 CAS9 . Following confirmation of successful integration of our CAS9 construct at the correct genomic location, CAS9 expression was confirmed using qRT-PCR (Fig 1B).
Expressing Cas9 in C. neoformans does not influence growth or virulence factor production For CRISPR to be useful in C. neoformans, CAS9 expression must not influence normal growth or any virulence-associated characteristics, such as growth at body temperature or production of melanin, urease, phospholipase, protease and capsule. Growth of the CAS9 strain H99 CAS9 was indistinguishable from wild-type at 30°C or at 37°C (human body temperature) on both rich YPD and synthetic YNB media (Fig 2A). Strains were again indistinguishable for melanin production as measured on L-DOPA medium, producing equivalent pigmentation at both temperatures ( Fig 2B). Indeed, there was no measurable distinction between the wild-type and CAS9 strains in any of the C. neoformans virulence factor in vitro assays performed at either 30 or 37°C; H99 and H99 CAS9 were the same for urease production on Christensen's agar, phospholipase production on egg yolk agar, protease production on BSA agar (Fig 2C), and capsule production in RPMI media (Fig 2D). In short, we saw no evidence that TEF1 promoter-driven heterologous expression of the humanized, SV40 NLScontaining S. pyogenes CAS9 construct influenced C. neoformans growth or virulence factor production in any way.

The C. neoformans Cas9 strain is indistinguishable from wild-type in growth or virulence
Expressing Cas9 in C. neoformans does not influence the progression of disease Given that the CAS9-expressing H99 CAS9 strain showed no detectable differences from wildtype in any in vitro analyses, we progressed our studies to animal infections. Mice infected intranasally with the CAS9 strain were indistinguishable from mice infected with wild-type, with both cohorts succumbing to infection within the typical timespan of three weeks (Fig 3A). To gain a deeper insight into the progression of infection, and to determine whether the presence of the CAS9 construct alters the tropism of the pathogen, a post-mortem analysis of fungal organ burden of the brain, lungs, kidneys, liver and spleen was performed. Again, the H99 CAS9 strain was indistinguishable from its wild-type parent (Fig 3B). Together with the in vitro data, these results indicated that Cas9 does not influence growth or virulence, and could therefore be employed in molecular studies of C. neoformans pathogenesis.

Creation of gRNA constructs designed to target ADE2
To function, CRISPR requires the expression of a nuclease and the synthesis of an RNA molecule to guide the nuclease activity. To investigate the utility of our CAS9 construct, we therefore needed to generate appropriate guide RNAs. However, transcribing these molecules in eukaryotes is problematic; most promoters are transcribed by RNA polymerase II, which performs 5' capping and polyadenylation that abolishes gRNA function. One solution is the use of an RNA polymerase III promoter such as U3, U6 or SNR52, however these add a 5' guanine to the transcript, limiting usable PAMs to those that have a corresponding G at the 5' end of the target sequence [37]. Furthermore, the precise start point of transcription must be known, and to our knowledge no RNA polymerase III promoters have been well characterized in C. neoformans.
The difficulties associated with eukaryotic gRNA synthesis have been addressed by Gao and Zhao [45], who devised a strategy where an RNA polymerase II promoter is used to transcribe a synthetic gene consisting of the Hammerhead and Hepatitis Delta Virus ribozymes flanking the desired gRNA sequence. Following transcription, the 5' capped, polyadenylated transcript undergoes self-cleavage, liberating an unmodified gRNA molecule. We followed this scheme, employing the ACT1 promoter / TRP1 terminator combination already used in a variety of C. neoformans constructs [35,46,49]. For the target sequence, we chose the wellcharacterized phosphoribosylaminoimidazole carboxylase-encoding ADE2 gene as a proof of principle example; loss of ADE2 results in the easily observable phenotypes of adenine auxotrophy and pink coloration due to accumulation of the purine biosynthetic intermediate Pribosylaminoimidazole [50].
Searching both strands of the ADE2 gene, we identified 275 canonical PAM sequences (NGG). To eliminate less favorable gRNA targets, we adopted the guidelines of Wang and colleagues [51], rejecting the 148 (54%) on the transcribed strand, and a further 41 (15%) where the PAM sequence was followed by a guanine residue; both of these characteristics have been shown to reduce gRNA effectiveness. Of the remaining 82, those with target sequences well within the suggested 40-80% GC content were prioritized [52]. Based on these criteria, we chose two targets, one approximately a third of the way through the ADE2 gene (gRNA1) (448 bp from the ATG), and a second approximately two thirds of the way through (gRNA2) (1256 bp from the ATG) (Fig 4).
The final engineered gRNA constructs contain the ACT1 C. neoformans promoter, an ADE2 pre-target region of 6 bp required for Hammerhead ribozyme function, the Hammerhead ribozyme, the 20 bp target sequence, the conserved non-target elements of the gRNA molecule, the HDV ribozyme and lastly the TRP1 terminator. We subcloned these products into the pSDMA25 Safe Haven vector for targeting to chromosome 1. The Safe Haven constructs were then integrated into the genome of C. neoformans type strain H99 to create strains H99 gRNA1 and H99 gRNA2 . Co-transformation with plasmid-borne CAS9 and gRNA constructs does not enhance ADE2 deletion frequency The Cas9/gRNA complex introduces double stranded breaks at the gRNA target sequence, which triggers activation of repair mechanisms and enhances homologous recombination. To determine if this process could increase the rate of targeted gene deletion in C. neoformans, we biolistically co-transformed an ade2::NEO deletion construct with a combination of CRISPR constructs on circular plasmids-either CAS9 plus gRNA1, or CAS9 plus gRNA2. By employing G418 media, there was no selective pressure to integrate or maintain the CRISPR plasmids; furthermore, unlike C. deneoformans, C. neoformans is unable to stably maintain episomal constructs, typically losing constructs within a few generations [7,9].
In our control transformation with just the ade2::NEO construct, the frequency of gene deletion was 33% (Fig 5), consistent with our previous investigations that have shown that this locus is more amenable than most in C. neoformans to deletion (35). Importantly, the co-transformation of the ade2::NEO construct with both the Cas9 plasmid and a gRNA plasmid showed no significant change in integration frequency, with the ADE2 deletion construct integrating at the correct position 30% of the time for gRNA1 and 26% for gRNA2 (Fig 5).

CRISPR in C. neoformans leads to increased homologous integration rates
Given that the inclusion of our circular, unselected CRISPR plasmids during co-transformation had no influence on the rate of ADE2 deletion, we next investigated the effect of having each of these elements stably expressed in the genome from the Safe Haven site during transformation with the ade2::NEO construct. The rate of ADE2 deletion was equivalent in H99, H99 CAS9 , H99 gRNA1 and H99 gRNA2 as expected; with only one of the elements of the CRISPR system present, no increase was expected (Fig 6A). Moreover, we saw no evidence of negative effects from stable expression of single CRISPR elements on the transformation process.
Once we had established that each CRISPR element alone did not influence the rate of ADE2 deletion, we employed the Safe Haven H99 CAS9 strain in co-transformations with the ade2::NEO and gRNA constructs. Unlike the previous experiments, we observed a substantial increase in the frequency of deletion of ADE2, raising to 70% (Fig 6B); under appropriate conditions, CRISPR functions in C. neoformans to enhance the rate of gene deletion without requiring integration of the gRNA plasmid (S1 Fig). Given our CRISPR success in H99 CAS9 , we next investigated the effect of employing the Safe Haven H99 gRNA1 and H99 gRNA2 strains in co-  transformations with the ade2::NEO and CAS9 constructs. In stark contrast to the H99 CAS9 success, the frequency of ADE2 deletion was not statistically different to the H99 control in these experiments ( Fig 6C). Together, these data confirm that we have successfully developed a functioning CRISPR system in C. neoformans by combining a strain stably expressing CAS9 with transient expression of self-cleaving gRNA constructs.

Discussion
The current tools and methodologies available for study of C. neoformans have significantly advanced our understanding of the mechanisms influencing virulence in this important pathogen. However, routine gene deletions can still be difficult to achieve, significantly hindering molecular genetic studies. Gene deletion through homologous recombination can be improved by several already published approaches. The low deletion rates already achieved are dependent on the use of homologous flanks in the transforming construct of around 1 kb, much longer than the size required in S. cerevisiae. Increasing these further may increase integration rates, however this benefit would be offset by introducing difficulties in PCR amplification and cloning [53]. Another approach is to use kuΔ strains; while they can increase the rates of recombination to nearly 100% [6,7], more recent publications have indicated that the Ku proteins may be playing a role during infection of a human host, and the loss of this protein results in altered virulence in mice [10,11]. These data indicate that it could be problematic to use Ku mutants for creating gene deletions for the study of pathogenesis in this species, and as a result these strains are rarely utilized. Unlike the strategies described above, CRISPR avoids interrupting NHEJ, as the Cas9 nuclease is directed to its target with a guide RNA. In this study, we have specifically demonstrated the ability of this system in the context of the ADE2 locus, increasing homologous integration rates to 70%. In the case of our constructs, this increase in homologous recombination was seen only when Cas9 was stably expressed following integration of the cassette into the genome, not when it was introduced on a non-integrative plasmid. We hypothesize that this is likely due to the construct not expressing the Cas9 nuclease to sufficient levels in time to influence the fate of transforming DNA before the plasmid is lost by the cell.
Importantly, and in stark contrast to recently published work [33], we have found no evidence that constitutive expression of Cas9 in C. neoformans affects normal growth, in vitro production of virulence factors, or the progression of disease in a murine model. However, there were some marked differences in the experimental approaches we employed which may explain this discrepancy. Wang and colleagues utilized the human codon-optimized CAS9 construct of Cong et al. [47], whilst we used the equivalent construct of Mali et al. [37]; these only show 81.1% nucleotide identity, and it is possible that the transcript of the Cong construct may contain cryptic intron splice sites recognized by C. neoformans, or the choice of codons may be highly disfavoured. Furthermore, the C. deneoformans construct employed a bovine growth hormone terminator, which may function differently in C. neoformans, and a different promoter, which may express Cas9 at a level that makes it toxic to this species. Finally, the C. deneoformans Cas9 is fused not only to the SV40 NLS, but to a triplicate FLAG tag and the nucleoplasmin NLS as well. While FLAG has been used in C. neoformans by others [54,55], we could find no record of the nuceloplasmin NLS being employed; perhaps it is this element that is toxic. We cannot be certain which of these explanations, if any, are the reason for the differences in this work and that published for C. deneoformans. Whatever the case, Wang and colleagues obtained excellent results with their construct in C. deneoformans, just as we have with our own construct in C. neoformans.
The CRISPR strain described here now joins the kuΔ mutants in providing researchers with a second potentially useful background that can be used to facilitate molecular genetic alterations in C. neoformans. The kuΔ strains are straightforward to use, and can be easily transformed with the construct of choice; however, the phenotype associated with the ku80Δ mutation is continually present unless outcrossed. The H99 CAS9 strain is more difficult to work with, requiring the creation of a gRNA in addition to the construct being transformed, but this background has the benefit of being an otherwise wild-type strain, with the heterologously expressed S. pyogenes nuclease inactive unless an appropriate guide molecule is provided. While untested here, we also postulate that this system could be used not only for creating gene deletions by providing a construct but also through the introduction of mutations in the absence of a template.
In summary, we have demonstrated the first proof of principle that CRISPR can be employed as a tool for high efficiency gene disruption in C. neoformans, holding significant potential for progressing the state of genetic research in this organism and consequently understanding its pathogenicity.
Supporting Information S1 Fig. Southern blot confirming absence of gRNA plasmids in strains generated through co-transformations employing the ade2::NEO construct with a gRNA plasmid. All strains were digested with PstI, and probed with the ACT1 promoter and TRP1 terminator regions common to the ACT1 and TRP1 loci, the gRNA plasmid, the hyg resistance cassette and the G418 resistance cassette. The expected bands as follows: ACT1 promoter in the native location, 5,715 bp; ACT1 and TRP1 associated with the HYG resistance marker for integration of CAS9: 2,555 and 4,762 bp; ACT1 and TRP1 associated with the NEO marker for ade2Δ: 3,756 and 1,365 bp. If the gRNA plasmid was present, a single additional 4,053 bp would be expected; only one strain showed unexpected bands, however these are likely associated with abnormal integration of the ade2::NEO construct at the ADE2 locus, as the strain was an adenine auxotroph, and obtaining unusual transformants is common when biolistically transforming C. neoformans. (TIF) S1