Whole-genome sequencing reveals highly specific gene targeting by in vitro assembled Cas9-ribonucleoprotein complexes in Aspergillus fumigatus

Whole-genome sequencing reveals highly specific gene targeting by in vitro assembled... Background: CRISPR/Cas9‑ based genome editing is quickly becoming a powerful tool within the field of fungal genetics. Adaptation of CRISPR/Cas9 systems are allowing for rapid and highly efficient gene targeting within fungi. We recently reported the adaptation of a simple CRISPR/Cas9 system for gene deletion that is effective across multiple genetic backgrounds of Aspergillus fumigatus. This system employs in vitro assembly of Cas9 ribonucleoproteins (RNPs) coupled with micro‑ homology repair templates for gene deletion. Although highly efficient at gene targeting in wild type genetic backgrounds of A. fumigatus, the potential for our system to produce unwanted off ‑ target muta‑ tions has not been addressed. Results: Next‑ generation Illumina sequencing was used to identify genome mutations among transformants isolated from standard (no Cas9) and Cas9‑ mediated integration of a hygromycin deletion cassette. Two different con‑ centrations of Cas9 were utilized to examine the association of Cas9 concentration with total numbers and types of genomic mutations. For each of the three test groups (zero, low, and high Cas9), three transformants were sequenced and compared to the parent strain. Bioinformatics analyses revealed the average number of total mutations to be sim‑ ilar among all three test groups. A. fumigatus transformation using standard, non‑ Cas9‑ mediated methods resulted in an average of 373 ± 28 mutations. In comparison, transformation with in vitro assembled Cas9‑ RNPs using either high (1 µg/µl) or low (0.5 µg/µl) levels of Cas9 resulted in an average of 326 ± 19 and 395 ± 69 mutations, respectively. In all cases, the vast majority of mutations identified were intergenic. No correlation between the amount of Cas9 utilized for transformation and the overall number of mutations was found. Finally, the specific type of mutation introduced during the transformation process was not Cas9‑ dependent, as both single‑ nucleotide polymorphisms and insertion/ deletion events were not significantly different between the experimental groups. Conclusions: CRISPR/Cas9‑ based genome editing in A. fumigatus using in vitro assembled RNPs coupled with micro‑ homology templates is a reliable method of gene targeting. This system is highly efficient and is not associated with increased off ‑ target mutations caused by introduction of the Cas9 nuclease. Keywords: Aspergillus fumigatus, CRISPR/Cas9, Genome editing, Off ‑ target mutation Background requires gene-targeting cassettes that contain ≥ 1000 Gene deletion in A. fumigatus wild type strains is plagued base pairs of homology to be cloned upstream and by low homologous recombination rates and typically downstream of a selection marker. The problem of low homologous recombination rates can be circumvented by using A. fumigatus strains mutated to have defective *Correspondence: jfortwen@uthsc.edu non-homologous end joining (NHEJ) DNA repair path- Department of Clinical Pharmacy and Translational Science, University ways [1, 2]. Although these strains have increased gene of Tennessee Health Science Center, Memphis, TN, USA © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Al Abdallah et al. Fungal Biol Biotechnol (2018) 5:11 Page 2 of 8 targeting efficiencies, deletion cassettes still require 500– nuclease and/or gRNA components or on building DNA- 1000  bp regions of flanking homology and the defective based constructs for expression of these components [7, NHEJ pathway(s) should be restored to ensure subtle 12]. genetic interactions between the targeted and NHEJ loci We have reported the adaptation of a CRISPR/Cas9 do not complicate phenotype interpretation. Finally, gene editing system that utilizes in vitro assembled Cas9- standard gene deletion methods in either wild type or RNPs coupled with microhomology repair templates NHEJ-defective backgrounds also rely on either multi- [13]. Rather than genetically altering strains to express step cloning or overlap extension PCR techniques to Cas9 or gRNAs, in vitro assembly relies on generating the build gene-targeting cassettes. Cas9 RNPs in a test tube before introducing them, along To improve gene targeting and genome editing in with a repair template (if required), into cells prepared A. fumigatus, CRISPR/Cas9 gene editing technology for transformation. In this system, the gRNA was formed has recently been implemented. In CRISPR-mediated in vitro by incubating a mixture of equal molar amounts genome editing, an RNA-directed Cas9 DNA nucle- of crRNA and tracrRNA until a complex is formed. The ase is employed to recognize and cleave specific DNA crRNA and tracrRNA are purchased separately and then sequences after forming a ribonucleoprotein (RNP) assembled into a gRNA complex so that the crRNA can complex with a guide RNA (gRNA) [3]. This gRNA is a be re-designed for each new protospacer the user desires duplex that is composed of a CRISPR RNA (crRNA) to target. Next, purified Cas9 enzyme was mixed with the and a transactivating CRISPR RNA (tracrRNA) [3]. crRNA-tracrRNA complexes and incubated to allow for The crRNA contains a 20-base region designated as the the formation of Cas9 RNPs. Cas9 concentrations of 1 “protospacer”, which guides specific DNA cleavage by and 0.5 µg/µl were utilized based on optimization experi- binding to the complementary protospacer in the tar- ments in our laboratory. The Cas9 RNP complexes were get genome [4]. However, Cas9-mediated DNA cleavage then used for standard transformation of A. fumigatus occurs only if the protospacer is followed by an “NGG” protoplasts along with 2 μg of hygromycin resistance cas- protospacer adjacent motif (PAM) in the target genome sette that is flanked by 35 base pair homology regions as [5]. Several CRISPR/Cas9 systems have been developed a repair template. The major advantages of this in  vitro in Aspergillus species [6–11]. In most of these organisms, assembly system are the simplicity (i.e., does not require the Cas9 enzyme and gRNA are introduced via expres- strain construction) and the potentially portability from sion constructs that are either contained within autono- strain to strain. Our system generated nearly 100% gene ΔKu80 mously replicating plasmid or are integrated into the targeting in the ΔakuB mutant, increased gene dele- genome. Those that employ plasmids, control Cas9 activ - tion frequencies in the wild type strain Af293 from the ity through the presence or absence of selective agents typical ~ 5% up to ~ 74%, and produced gene targeting in the medium for plasmid maintenance whereas those efficiencies of ~ 90% in a clinical isolate [13]. Although designed for integration typically rely on regulatable pro- gene targeting was greatly improved by our method, and moters. CRISPR/Cas9 systems are becoming ever more is generally highly efficient in CRISPR/Cas9 methods sophisticated in Aspergillus, as evidenced by recent work developed for other the Aspergillus species, the potential showing that highly-efficient marker-free gene editing for off-target mutations has largely not been addressed. can be accomplished in A. nidulans, A. niger, and A. ory- Multiple recent studies have highlighted the potential zae strains defective in NHEJ [11]. In addition, highly effi - for CRISPR/Cas9-based gene editing systems to induce cient multi-site targeting is now possible in each of these unwanted off-target mutations [14–16]. This is typically Aspergillus species [11]. In A. fumigatus, the original believed to occur by promiscuous induction of dou- use of CRISPR/Cas9 involved strains that constitutively ble strand DNA breaks at sites other than the intended expressed Cas9 from an integrated construct [7]. A later protospacer. To ensure the reliability of our in  vitro iteration in A. fumigatus employed autonomously repli- assembly system, we sought to examine the impact of cating plasmids for Cas9 and gRNA expression and also exogenous Cas9 addition during the transformation pro- utilized selectable marker cassettes (“repair templates” in cess on the induction of genomic mutations in A. fumiga- CRISPR/Cas9 terminology) with microhomology regions tus using transformants from the studies we previously that are only 35–50  bp in length, showing that efficient performed in the Af293 genetic background. gene targeting can be accomplished in A. fumigatus with only small regions of DNA homology [12]. Although Results CRISPR/Cas9 gene targeting appears highly efficient To examine the potential for off-target mutations induced in A. fumigatus, the systems in place thus far rely on by the transient presence of gRNAs and the Cas9 nucle- genetically altering strains to express the required Cas9 ase in our system, we completed next-generation whole- genome sequencing on a subset of transformants isolated Al Abdallah et al. Fungal Biol Biotechnol (2018) 5:11 Page 3 of 8 from our previous study and performed a comparative allowed for colony color-based identification of homolo - analysis of the relative numbers and characteristics of gous integrations. The major difference between the sec - genomic mutations. In our previous study, reporting the ond and third experiments of our previous study was the optimization of in  vitro assembled Cas9 RNPs for gene use of either high (1 µg/µl) or low (0.5 µg/µl) concentra- targeting in A. fumigatus, we generated multiple transfor- tions of the Cas9 nuclease, respectively. Therefore, the mants from three basic experimental designs. In the first, experimental groups for our whole genome analyses in we utilized the wild type reference isolate, Af293, and the current study included three isolates from each of performed a transformation using standard protoplasting these three transformation conditions: no Cas9 (stand- protocols to ectopically integrate a hygromycin selection ard transformation protocol), 0.5  µg/µl Cas9 and 1  µg/ cassette, hygR [13]. In the second and third experiments, µl Cas9 (Table  1). As a reference, we also sequenced the we used Af293 to perform targeted CRISPR/Cas9-medi- parent strain, Af293. A. fumigatus Af293 is the genome ated integrations of the same hygR cassette to delete the reference isolate with a well-annotated genome. How- coding region of a polyketide synthase, pksP [13]. The ever, this parent isolate was re-sequenced to account pksP locus was chosen for protocol optimization as it for genomic mutations that may have arisen during the repetitive sub-culturing of this strain in our laboratory. Average genome coverage ranged from 38× to 62× for all isolates (Table 2). Table 1 Strains used in this study Comparative bioinformatics analyses revealed that Strain name Background multiple genomic mutations were present in the transfor- mants of all strains, regardless of the presence or absence Af293 Wild type of Cas9. In the experimental group lacking Cas9, a group NC1 hygR (no Cas9) average of 373 ± 28 mutations were identified (Table  2). NC2 hygR (no Cas9) Considered alone, this finding demonstrates the potential NC3 hygR (no Cas9) mutagenic nature of A. fumigatus protoplast transforma- HC4 ΔpksP‑hygR (1 µg/µl Cas9) tion. This standard transformation protocol may induce HC5 ΔpksP‑hygR (1 µg/µl Cas9) intense, albeit temporary, cellular stress as it requires the HC6 ΔpksP‑hygR (1 µg/µl Cas9) enzymatic digestion of the cell wall to release membrane- LC7 ΔpksP‑hygR (0.5 µg/µl Cas9) bound protoplasts followed by recovery on an osmoti- LC8 ΔpksP‑hygR (0.5 µg/µl Cas9) cally stabilized agar medium. Transformants from both LC9 ΔpksP‑hygR (0.5 µg/µl Cas9) the high (1 µg/µl) and low (0.5 µg/µl) Cas9 concentration For strains names, NC = no Cas9, HC = high Cas9, and LC = low Cas9. Indicated experiments displayed total numbers of genomic muta- at the right are the concentrations of Cas9 used to produce each mutant strain. hygR = hygromycin resistance cassette; pksP = polyketide synthase; ΔpksP- tions similar to the no Cas9 control, with an average of hygR = pksP locus replaced by hygR Table 2 Cas9-mediated gene deletion is not associated with increased genomic mutations in A. fumigatus NC1 NC2 NC3 HC4 HC5 HC6 LC7 LC8 LC9 Average coverage 53× 62× 60× 58× 54× 61× 56× 49× 38× Total mutations 396 342 385 345 326 307 345 366 474 Average total muta‑ 373 ± 28 SD 326 ± 19 SD (p > 0.05) 395 ± 69 SD (p > 0.05) tions Analysis based on type of mutation identified SNPs 371 318 363 321 314 292 326 347 439 Indels 25 24 22 24 12 15 19 19 35 Analysis based on location of mutation Intergenic 380 (96%) 331 (97%) 379 (98%) 331 (96%) 320 (98%) 301 (98%) 339 (98%) 359 (98%) 446 (94%) Average intergenic 363 ± 28 SD 317 ± 15 SD (p = 0.04) 381 ± 57 SD (p > 0.05) Coding region 16 (4%) 11 (3%) 6 (2%) 14 (4%) 6 (2%) 6 (2%) 6 (2%) 7 (2%) 28 (6%) Average coding 11 ± 5 SD 9 ± 5 SD (p > 0.05) 14 ± 12 SD (p > 0.05) region Displayed are the total and average number of mutations among the three experimental groups: no (“NC”—0 µg/µl), low (“LC”—0.5 µg/µl) and high (“HC”—1 µg/ µl) levels of Cas9. For the intergenic and coding region mutation rows, the numbers in parentheses represent the percent (%) of total mutations. For the average mutations per group, the mean ± standard deviation (SD) is provided. The Student’s t test assuming unequal variance was used for statistical comparisons of the Cas9 (HC or LC) and the no-Cas9 (NC) groups and p values are presented Al Abdallah et al. Fungal Biol Biotechnol (2018) 5:11 Page 4 of 8 326 ± 19 and 395 ± 69, respectively (Table  2). Among all total mutations) than those from the high concentration transformants, only a small subset of the identified muta - Cas9 experiment (~ 0.5% of total mutations). Thus, we tions were located within coding regions of the genome, interpret these mutations not as a consequence of Cas9 as the vast majority (> 96% for all isolates) were intergenic presence but as variability among isolates resulting from (Table 2). A single transformant (LC9) from the low Cas9 the standard protoplast transformation process. concentration experiment was found to contain relatively Cas9-induced double strand breaks (DSBs) can be higher levels of total mutations, with a total of 474, and repaired by two major pathways in the cell: the NHEJ a slightly increased distribution of these mutations into DNA repair pathway or homology directed repair [17]. coding regions (~ 5.9% of total compared to an average The NHEJ repair pathway is error prone and induces of ~ 2.6% for all other transformants). However, as this small insertion and deletion (indel) events into the transformant is an isolate from the low Cas9 concentra- genome [17]. Therefore, a negative consequence of Cas9- tion experiment, this apparent increase in total number mediated gene editing could be the promiscuous induc- of mutations and distribution towards coding regions is tion of DSBs resulting in increased indels throughout the not likely associated with the activity of Cas9. Further affected transformant(s). To see if this specific type of supporting this assertion, we found no statistically signif- off-target mutation may exist in our collection, we also icant difference in the total number of mutations identi - analyzed transformants from each experimental group fied among the experimental groups (Table  2). Therefore, for the disproportional generation of indels versus sin- our analyses revealed no Cas9 concentration-dependent gle nucleotide polymorphisms (SNPs). To do this, the increase in induction of genomic mutations. The muta - total genomic mutations (intergenic and coding region) tions identified in our study are likely induced by the were classified into either SNPs or indels and the result - transformation process, including cell wall digestion. ing numbers were averaged for each experimental group. To further examine if the presence of Cas9 during Our data indicated that the generation of SNPs was transformation may influence the type of mutation intro - favored over indels (1–80 nucleotides in length) within duced, we analyzed additional characteristics of the iden- each experimental group and no significant differences tified genomic variations. Those mutations identified in the relative amounts of either mutation were noted within coding regions displayed comparable distribu- (Fig. 1). When expressed as a percent of total mutations, tions among the categories defined in Table  3. In general, the non-Cas9 mediated transformation generated strains most mutations were located within the 3′ UTRs of genes containing 6.4% indels, whereas the high and low con- regardless of the presence or absence of Cas9. The only centration Cas9 transformants had 5.2 and 6.2% indels, discrepancy noted was that all transformants within the respectively. Therefore, the concentration of Cas9 is not high and low concentration Cas9 groups displayed a low associated with a disproportional increase in the num- number of non-synonymous mutations identified within ber of indels between experimental groups. Although coding regions, whereas the standard non-Cas9 transfor- our protospacers were designed to have minimal off-site mants were free of non-synonymous mutations (Table 3). complementarity, some level of similarity between our However, similar to our findings with total numbers of designed sequences and distant areas of the genome is mutations, the low concentration Cas9 transformants unavoidable. To ensure that off-target complementarity contained more non-synonymous mutations (~ 0.8% of of our gRNA complexes was not promiscuously driving Table 3 Cas9-mediated gene deletion does not cause alterations in the types of coding region mutations in A. fumigatus 0 µg/µl Cas9 1 µg/µl Cas9 0.5 µg/µl Cas9 NC1 NC2 NC3 HC4 HC5 HC6 LC7 LC8 LC9 3′ UTR 9 9 2 9 2 – 1 1 10 5′ UTR 3 – 1 2 – 1 1 – 1 Frameshift 1 – – 1 – – – – – Intron 2 2 2 1 1 1 2 3 2 Non‑synonymous – – – 1 1 3 2 1 6 Start lost – – – – 1 – – – – Synonymous – – – – 1 1 – 1 9 Splice region 1 – 1 – – – – 1 – Shown are the numbers and types of identified mutations located within coding regions of transformants within the three experimental groups Al Abdallah et al. Fungal Biol Biotechnol (2018) 5:11 Page 5 of 8 CRISPR/Cas9 components, this system can be utilized to study gene and pathway function across many isolates. Multiple studies have indicated that CRISPR/Cas9-based gene editing is potentially associated with off-target mutations whereas many studies have also described this system as highly specific [18]. Off-target mutations could be due to promiscuous activity of the Cas9 nucle- ase, yet more often seem to be caused by non-specific binding of gRNA [18]. Using the system we have adopted for A. fumigatus, both of these potential issues are prop- erly addressed [13]. The Cas9 nuclease is introduced as a purified protein and, therefore, is only transiently pre - sent. Compared to systems that constitutively or condi- tionally express Cas9, this should reduce the potential for unwanted mutagenesis of the genome generated by pro- miscuous Cas9 activity. In fact, in a few yeast and para- site studies, expression of CRISPR/Cas9 components was associated with toxicity [19–21]. Our system avoids this problem entirely. Additionally, using the rules for PAM selection, protospacer design, and RNP assembly that we have previously reported [13], we were able to generate Cas9-transformants that did not show increased genomic Fig. 1 Cas9‑mediated gene deletion does not cause a dispropor ‑ variability. Because the genomic variation identified in tional increase in SNPs or indels in A. fumigatus. Segregation of genomic mutations into SNPs (b) and indels (a) revealed that the our study is not correlated with CRISPR/Cas9 methods, concentration of Cas9 was not positively associated with an increase our data also suggest that either the single sub-culture in a specific subset of mutation. Shown are the average number of of a strain of A. fumigatus within the laboratory or the mutation events within each experimental group. Student’s t test stress induced by our standard protoplast transformation assuming unequal variance was utilized for statistical comparison procedure induces genomic variation. between the “no Cas9” and either the low (0.5 µg/µl) or high (1 µg/µl) Cas9 group Although whole genome sequencing of transformants has proven useful for the analysis of off-target effects in many systems, this technique does have the limitation of being unable to identify Cas9-induced DSBs that are per- Cas9 to generate unwanted DSBs and subsequent off-tar - fectly repaired. With this caveat in mind, whole genome get mutations, we finally interrogated each transformant sequencing has been successfully applied to characterize genome for variations surrounding ten potential off- the potential for off-target effects of Cas9-based transfor - target sites. These off-target sites were defined as areas mations in the plant pathogen Ustilago maydis [22]. This where the protospacer had twelve or more base pairs of study relied on expressing Cas9 from a strong constitutive complementarity. For both the 5′ and 3′ protospacer, our promoter on a self-replicating plasmid. After transforma- analysis found zero genomic mutations in these areas tion, strains could then be cured of this plasmid to avoid (data not shown). Together, these data indicate that continuous passage and growth in the presence of Cas9. CRISPR/Cas9 editing by our methods is highly specific in Whole genome sequencing of transformants acquired A. fumigatus. in this study revealed that none of the identified genome mutations were likely to be due to Cas9 activity mediated Discussion by gRNA binding [22]. Our results support the same con- The implementation of novel gene editing technolo - clusion when purified Cas9 is added exogenously to A. gies in A. fumigatus, like CRISPR/Cas9, is a critical step fumigatus. It is of note, however, that our study only inves- toward making significant advances in studies related to tigated two concentrations of Cas9, did not examine the virulence and antifungal drug resistance in this impor- effects of varying amounts of crRNA or tracrRNA or the tant human pathogen. We have shown that our system, ratios of Cas9 to tracrRNA and crRNA, and only inves- relying on in vitro assembled Cas9-RNP complexes, pro- tigated off-target effects upon targeting of only one gene. duces highly efficient gene targeting in multiple genetic It is possible that by targeting a different protospacer or backgrounds of A. fumigatus [13]. Because it does not by significantly altering the Cas9 RNP composition, a dif - rely on strains that have been genetically engineered to ferent outcome might have been observed. However, even increase homologous recombination rates or to express Al Abdallah et al. Fungal Biol Biotechnol (2018) 5:11 Page 6 of 8 though off-target effects might increase under these other resuspended in 800  μl of buffer AP1 and 8  μl of RNase conditions, further efforts that try to minimize them A and vigorously vortexed. Following 3  h of incubation can be pursed. For example, bioinformatic tools are now at 65  °C, the fungal lysate was centrifuged for 5  min at available to interrogate genomes for potential off-target 20,000×g and the supernatant was transferred to a new sites, including Cas-OFFinder and Cas-Designer [23, 1.5  ml tube. Next, 260  μl of buffer P3 were added to the 24]. Databases like these can aid in protospacer selection supernatant and the mixture was incubated for 5 min on and gRNA design to minimize the potential for off-target ice, then centrifuged for 5  min at 20,000×g. The super - mutations when targeting a new genomic locus. Also, natant was then transferred to QIAshredder spin column multiple studies into ways to minimize off-target muta - (700 μl at a time) and centrifuged for 2 min at 20,000×g. tions in CRISPR/Ca9 systems have been published and The flow-through was collected into a 2 ml tube without new techniques are constantly being pursued. One tech- disturbing the pellet, and 1.5 volumes of buffer AW1 was nique might be to limit the time of Cas9 activity in the cell added and mixed by pipetting. The mixture was trans - via use of photoactivatable split-Cas9 [25]. Our system ferred (700 μl at a time) into a DNeasy Mini spin column accomplishes limited Cas9 activity through introduction and the genomic DNA was allowed to bind to the col- of the Cas9 enzyme. However, if required, the specificity umn membrane by centrifuging for 1  min at ≥ 6000×g. of our system could also be further bolstered by employ- The genomic DNA was washed first with 700 μl of AW2 ing high-fidelity or rationally engineered Cas9 enzymes buffer and centrifuged for 1 min at ≥ 6000×g, followed by with increased specificity or through the use of truncated a second washing step using 300  μl of Buffer AW2 and versions of single-gRNAs [26–28]. centrifugation at 20,000×g for 5  min. The second wash - ing step was essential to remove any residual ethanol Conclusions on the membrane before elution step. The spin column The data provided here demonstrate that CRISPR/Cas9- was transferred to a new 1.5  ml tube. For the elution of mediated gene targeting, using our in  vitro assembled genomic DNA, 100  μl Buffer 5  mM Tris–HCl (pH 8.5) Cas9-RNP system, does not cause an increase in genomic was added to the center of the column and the column variation over standard transformation protocols. We was incubated at room temperature for 5  min, followed also identified no disproportional Cas9-dependent by a centrifugation step for 1  min at ≥ 6000×g. Final increase in SNPs or indels among treated strains. There - DNA concentrations were quantified using Nanodrop fore, Cas9-mediated gene deletion using in  vitro assem- and Qubit Fluorometer, following the manufacturer’s bled Cas9-RNPs coupled with microhomology repair protocol. templates is a reliable method for generating targeted mutations in A. fumigatus. Library preparation and bioinformatics analyses Library preparations and genome sequencing reactions Methods were performed at the University of Alabama at Birming- Strains and culture conditions ham Heflin Center for Genomic Science. The Qiagen Strains used for this study are listed in Table  1. Strain QIAseq FX DNA prep kit was used for library prepara- Af293 is the A. fumigatus reference genome isolate [29] tions, following the manufacturer’s instructions. Paired and all transformant strains employed for whole genome end 300 base pair sequencing reads were generated on sequencing were generated, as part of a previous study, the Illumina MiSeq following standard protocols. Bioin- in this genetic background [13]. All strains were main- formatics services were provided by code4DNA (www. tained on glucose minimal media (GMM) agar [30], sup- code4DNA.com). Reads from each sample were aligned plemented with hygromycin (150  µg/ml) for selection. using bwa mem (v0.7.15) to the A. fumigatus reference Conidia were harvested in water from three-day old genome build A_fumigatus_Af293_version_s03-m05-r05 plates and enumerated by hemocytometer. downloaded from AspGD.org [32]. Samtools (v1.3) fix - mate and rmdup were used to remove PCR duplicates Genomic DNA extraction [33]. Sequence mutations were called using FreeBayes Genomic DNA was extracted following a slight modifi - (v1.1.0) with the haploid population-based model [34]. cation of previous published protocols, using the Qia- Low quality (QUAL < 30) and low depth (DP < 10) muta- gen DNeasy Plant Mini Kit [31]. Briefly, strains were tions were filtered out using VCFtools (v0.1.15) which inoculated in GMM broth at a conidial density of 10 was also used to remove variant calls where no sequence conidia/ml and incubated for 20 h at 37 °C with shaking reads were available for at least one sample. The popu - at 250  rpm. Mycelia were harvested by vacuum filtra - lation.vcf file was split into individual samples using tion and 300 mg of a semi-dry mycelial mat was crushed VCFtools vcf-subset. Mutations were annotated using under liquid nitrogen. The resulting mycelial powder was snpEff (v4.3r) and VCFtools vcf-isec was used to select Al Abdallah et al. Fungal Biol Biotechnol (2018) 5:11 Page 7 of 8 10. Weyda I, Yang L, Vang J, Ahring BK, Lübeck M, Lübeck PS. A comparison of mutations found in each affected samples but not in the Agrobacterium‑mediated transformation and protoplast ‑mediated trans‑ Ref. [35]. formation with CRISPR–Cas9 and bipartite gene targeting substrates, as effective gene targeting tools for Aspergillus carbonarius. J Microbiol Methods. 2017;135:26–34. Abbreviations 11. Nødvig CS, Hoof JB, Kogle ME, Jarczynska ZD, Lehmbeck J, Klitgaard DK, CRISPR: clustered regularly interspaced short palindromic repeats; crRNA: Mortensen UH. Efficient oligo nucleotide mediated CRISPR–Cas9 gene CRISPR RNA; DSB: double strand break; GMM: glucose minimal media; gRNA: editing in Aspergilli. Fung Genet Biol. 2018. https://doi.org/10.1016/j. guide RNA; Indel: insertion/deletion; RNP: ribonucleoprotein; SNP: single‑ fgb.2018.01.004. nucleotide polymorphism; tracrRNA: trans‑activating crRNA. 12. Zhang C, Meng X, Wei X, Lu L. Highly efficient CRISPR mutagenesis by microhomology‑mediated end joining in Aspergillus fumigatus. Fung Authors’ contributions Genet Biol. 2016;86(Supplement C)):47–57. QAA performed the transformation of strains to obtain isolates for genome 13. Al Abdallah Q, Ge W, Fortwendel JR. A simple and universal system for sequencing and isolated genomic DNA for sequencing studies. WG cultured gene manipulation in Aspergillus fumigatus in vitro‑assembled Cas9‑ guide and prepared the selected strains for analysis. JRF contracted for the genomic RNA ribonucleoproteins coupled with microhomology repair templates. sequencing and bioinformatics analyses. QAA, WG, AMV, ACOS and JRF con‑ mSphere. 2017. https://doi.org/10.1128/msphere.00446‑17. tributed to data interpretation and manuscript preparation. All authors read 14. Zhang X‑H, Tee LY, Wang X ‑ G, Huang Q‑S, Yang S‑H. Off‑target effects in and approved the final manuscript. CRISPR/Cas9‑mediated genome engineering. Mol Ther Nucleic Acids. 2015;4:e264. https://doi.org/10.1038/mtna.2015.37. Acknowledgements 15. Schaefer KA, Wu W‑H, Colgan DF, Tsang SH, Bassuk AG, Mahajan VB. The authors would like to thank code4DNA (www.code4DNA.com) for con‑ Unexpected mutations after CRISPR–Cas9 editing in vivo. Nat Methods. tributing written summaries of how bioinformatics analyses were performed. 2017;14:547–8. 16. Zhang Q, Xing H‑L, Wang Z ‑P, Zhang H‑ Y, Yang F, Wang X‑ C, Chen Q‑ J. Competing interests Potential high‑frequency off‑target mutagenesis induced by CRISPR/Cas9 The authors declare that they have no competing interests. in Arabidopsis and its prevention. Plant Mol Biol. 2018;96:445–56. 17. Lieber MR. The mechanism of double‑strand DNA break repair by Ethics approval and consent to participate the nonhomologous DNA end joining pathway. Ann Rev Biochem. Not applicable. 2010;79:181–211. 18. O’Geen H, Yu AS, Segal DJ. How specific is CRISPR/Cas9 really? Cur Opin Funding Chem Biol. 2015;29:72–8. This work was supported by NIH Grant R01AI106925 to J.R.F. 19. Ryan OW, Skerker JM, Maurer MJ, Li X, Tsai JC, Poddar S, Lee ME, DeLoache W, Dueber JE, Arkin AP, Cate JHD. Selection of chromosomal DNA librar‑ ies using a multiplex CRISPR system. eLife. 2014;3:e03703. https://doi. Publisher’s Note org/10.7554/elife.03703. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 20. Generoso WC, Gottardi M, Oreb M, Boles E. Simplified CRISPR–Cas lished maps and institutional affiliations. genome editing for Saccharomyces cerevisiae. J Microbiol Methods. 2016;127:203–5. Received: 2 March 2018 Accepted: 18 April 2018 21. Jiang W, Brueggeman AJ, Horken KM, Plucinak TM, Weeks DP. Successful transient expression of Cas9 and single guide RNA genes in Chla- mydomonas reinhardtii. Eukaryot Cell. 2014;13:1465–9. 22. Schuster M, Schweizer G, Reissmann S, Kahmann R. Genome editing in Ustilago maydis using the CRISPR–Cas system. Fungal Genet Biol. 2016;89:3–9. References 23. Park J, Bae S, Kim J‑S. Cas‑Designer: a web ‑based tool for choice of 1. Krappmann S, Sasse C, Braus GH. Gene targeting in Aspergillus fumigatus CRISPR–Cas9 target sites. Bioinformatics. 2015;31:4014–6. by homologous recombination is facilitated in a nonhomologous end‑ 24. Bae S, Park J, Kim J‑S. Cas‑ OFFinder: a fast and versatile algorithm that joining‑ deficient genetic background. Eukaryot Cell. 2006;5:212–5. searches for potential off‑target sites of Cas9 RNA‑ guided endonucleases. 2. da Silva Ferreira ME, Kress MRVZ, Savoldi M, Goldman MHS, Härtl A, Bioinformatics. 2014;30:1473–5. Heinekamp T, Brakhage AA, Goldman GH. The akuBKU80 mutant defi‑ 25. Nihongaki Y, Kawano F, Nakajima T, Sato M. Photoactivatable CRISPR– cient for nonhomologous end joining is a powerful tool for analyzing Cas9 for optogenetic genome editing. Nat Biotechnol. 2015;33:755–60. pathogenicity in Aspergillus fumigatus. Eukaryot Cell. 2006;5:207–11. 26. Kleinstiver BP, Pattanayak V, Prew MS, Tsai SQ, Nguyen NT, Zheng Z, Joung 3. Sander JD, Joung JK. CRISPR–Cas systems for editing, regulating and JK. High‑fidelity CRISPR–Cas9 nucleases with no detectable genome ‑ targeting genomes. Nat Biotechnol. 2014;32:347–55. wide off‑target effects. Nature. 2016;529:490–5. 4. Shah SA, Erdmann S, Mojica FJM, Garrett RA. Protospacer recognition 27. Slaymaker IM, Gao L, Zetsche B, Scott DA, Yan WX, Zhang F. Ration‑ motifs. RNA Biol. 2013;10:891–9. ally engineered Cas9 nucleases with improved specificity. Science. 5. Mojica FJM, Díez‑ Villaseñor C, García‑Martínez J, Almendros C. Short motif 2016;351:84–8. sequences determine the targets of the prokaryotic CRISPR defence 28. Fu Y, Sander JD, Reyon D, Cascio VM, Joung JK. Improving CRISPR–Cas system. Microbiology. 2009;155:733–40. nuclease specificity using truncated guide RNAs. Nat Biotechnol. 6. Nødvig CS, Nielsen JB, Kogle ME, Mortensen UH. A CRISPR–Cas9 2014;32:279–84. system for genetic engineering of filamentous fungi. PLoS ONE. 29. Nierman WC, Pain A, Anderson MJ, Wortman JR, Kim HS, Arroyo J, Berri‑ 2015;10:e0133085. https://doi.org/10.1371/journal.pone.0133085. man M, Abe K, Archer DB, Bermejo C, Bennett J, Bowyer P, Chen D, Collins 7. Fuller KK, Chen S, Loros JJ, Dunlap JC. Development of the CRISPR/Cas9 M, Coulsen R, Davies R, Dyer PS, Farman M, Fedorova N, Fedorova N, system for targeted gene disruption in Aspergillus fumigatus. Eukaryot Feldblyum TV, Fischer R, Fosker N, Fraser A, García JL, García MJ, Goble A, Cell. 2015;14:1073–80. Goldman GH, Gomi K, Griffith‑ Jones S, Gwilliam R, Haas B, Haas H, Harris 8. Katayama T, Tanaka Y, Okabe T, Nakamura H, Fujii W, Kitamoto K, Maruy‑ D, Horiuchi H, Huang J, Humphray S, Jiménez J, Keller N, Khouri H, Kita‑ ama J. Development of a genome editing technique using the CRISPR/ moto K, Kobayashi T, Konzack S, Kulkarni R, Kumagai T, Lafton A, Latgé J‑P, Cas9 system in the industrial filamentous fungus Aspergillus oryzae. Li W, Lord A, Lu C, Majoros WH, May GS, Miller BL, Mohamoud Y, Molina Biotechnol Lett. 2016;38:637–42. M, Monod M, Mouyna I, Mulligan S, Murphy L, O’Neil S, Paulsen I, Peñalva 9. Zhang C, Meng X, Wei X, Lu L. Highly efficient CRISPR mutagenesis by MA, Pertea M, Price C, Pritchard BL, Quail MA, Rabbinowitsch E, Rawlins microhomology‑mediated end joining in Aspergillus fumigatus. Fung N, Rajandream M‑A, Reichard U, Renauld H, Robson GD, de Córdoba Genet Biol. 2016;86:47–57. SR, Rodríguez‑Peña JM, Ronning CM, Rutter S, Salzberg SL, Sanchez M, Al Abdallah et al. Fungal Biol Biotechnol (2018) 5:11 Page 8 of 8 Sánchez‑Ferrero JC, Saunders D, Seeger K, Squares R, Squares S, Takeuchi 32. Li H, Durbin R. Fast and accurate short read alignment with Burrows– M, Tekaia F, Turner G, de Aldana CRV, Weidman J, White O, Woodward J, Yu Wheeler transform. Bioinformatics. 2009;25:1754–60. J‑H, Fraser C, Galagan JE, Asai K, Machida M, Hall N, Barrell B, Denning DW. 33. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abe‑ Genomic sequence of the pathogenic and allergenic filamentous fungus casis G, Durbin R. The sequence alignment/map format and SAMtools. Aspergillus fumigatus. Nature. 2005;438:1151–6. Bioinformatics. 2009;25:2078–9. 30. Shimizu K, Keller NP. Genetic involvement of a cAMP‑ dependent protein 34. Garison E, Marth G. Haplotype‑based variant detection from short ‑read kinase in a G protein signaling pathway regulating morphological and sequencing. 2012. arXiv preprint: arXiv:1207.3907[q‑bio.GN]. chemical transitions in Aspergillus nidulans. Genetics. 2001;157:591–600. 35. Cingolani P, Platts A, Wang LL, Coon M, Nguyen T, Wang L, Land SJ, Lu X, 31. Hagiwara D, Takahashi H, Watanabe A, Takahashi‑Nakaguchi A, Kawamoto Ruden DM. A program for annotating and predicting the effects of single S, Kamei K, Gonoi T. Whole‑ genome comparison of Aspergillus fumigatus nucleotide polymorphisms, SnpE. F ff ly. 2012;6:80–92. strains serially isolated from patients with Aspergillosis. J Clin Microbiol. 2014;52:4202–9. Ready to submit your research ? Choose BMC and benefit from: fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Fungal Biology and Biotechnology Springer Journals

Whole-genome sequencing reveals highly specific gene targeting by in vitro assembled Cas9-ribonucleoprotein complexes in Aspergillus fumigatus

Free
8 pages
Loading next page...
 
/lp/springer_journal/whole-genome-sequencing-reveals-highly-specific-gene-targeting-by-in-dugGDuGsus
Publisher
BioMed Central
Copyright
Copyright © 2018 by The Author(s)
Subject
Life Sciences; Microbiology; Mycology; Biotechnology; Applied Microbiology; Microbial Genetics and Genomics; Plant Pathology
eISSN
2054-3085
D.O.I.
10.1186/s40694-018-0057-2
Publisher site
See Article on Publisher Site

Abstract

Background: CRISPR/Cas9‑ based genome editing is quickly becoming a powerful tool within the field of fungal genetics. Adaptation of CRISPR/Cas9 systems are allowing for rapid and highly efficient gene targeting within fungi. We recently reported the adaptation of a simple CRISPR/Cas9 system for gene deletion that is effective across multiple genetic backgrounds of Aspergillus fumigatus. This system employs in vitro assembly of Cas9 ribonucleoproteins (RNPs) coupled with micro‑ homology repair templates for gene deletion. Although highly efficient at gene targeting in wild type genetic backgrounds of A. fumigatus, the potential for our system to produce unwanted off ‑ target muta‑ tions has not been addressed. Results: Next‑ generation Illumina sequencing was used to identify genome mutations among transformants isolated from standard (no Cas9) and Cas9‑ mediated integration of a hygromycin deletion cassette. Two different con‑ centrations of Cas9 were utilized to examine the association of Cas9 concentration with total numbers and types of genomic mutations. For each of the three test groups (zero, low, and high Cas9), three transformants were sequenced and compared to the parent strain. Bioinformatics analyses revealed the average number of total mutations to be sim‑ ilar among all three test groups. A. fumigatus transformation using standard, non‑ Cas9‑ mediated methods resulted in an average of 373 ± 28 mutations. In comparison, transformation with in vitro assembled Cas9‑ RNPs using either high (1 µg/µl) or low (0.5 µg/µl) levels of Cas9 resulted in an average of 326 ± 19 and 395 ± 69 mutations, respectively. In all cases, the vast majority of mutations identified were intergenic. No correlation between the amount of Cas9 utilized for transformation and the overall number of mutations was found. Finally, the specific type of mutation introduced during the transformation process was not Cas9‑ dependent, as both single‑ nucleotide polymorphisms and insertion/ deletion events were not significantly different between the experimental groups. Conclusions: CRISPR/Cas9‑ based genome editing in A. fumigatus using in vitro assembled RNPs coupled with micro‑ homology templates is a reliable method of gene targeting. This system is highly efficient and is not associated with increased off ‑ target mutations caused by introduction of the Cas9 nuclease. Keywords: Aspergillus fumigatus, CRISPR/Cas9, Genome editing, Off ‑ target mutation Background requires gene-targeting cassettes that contain ≥ 1000 Gene deletion in A. fumigatus wild type strains is plagued base pairs of homology to be cloned upstream and by low homologous recombination rates and typically downstream of a selection marker. The problem of low homologous recombination rates can be circumvented by using A. fumigatus strains mutated to have defective *Correspondence: jfortwen@uthsc.edu non-homologous end joining (NHEJ) DNA repair path- Department of Clinical Pharmacy and Translational Science, University ways [1, 2]. Although these strains have increased gene of Tennessee Health Science Center, Memphis, TN, USA © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Al Abdallah et al. Fungal Biol Biotechnol (2018) 5:11 Page 2 of 8 targeting efficiencies, deletion cassettes still require 500– nuclease and/or gRNA components or on building DNA- 1000  bp regions of flanking homology and the defective based constructs for expression of these components [7, NHEJ pathway(s) should be restored to ensure subtle 12]. genetic interactions between the targeted and NHEJ loci We have reported the adaptation of a CRISPR/Cas9 do not complicate phenotype interpretation. Finally, gene editing system that utilizes in vitro assembled Cas9- standard gene deletion methods in either wild type or RNPs coupled with microhomology repair templates NHEJ-defective backgrounds also rely on either multi- [13]. Rather than genetically altering strains to express step cloning or overlap extension PCR techniques to Cas9 or gRNAs, in vitro assembly relies on generating the build gene-targeting cassettes. Cas9 RNPs in a test tube before introducing them, along To improve gene targeting and genome editing in with a repair template (if required), into cells prepared A. fumigatus, CRISPR/Cas9 gene editing technology for transformation. In this system, the gRNA was formed has recently been implemented. In CRISPR-mediated in vitro by incubating a mixture of equal molar amounts genome editing, an RNA-directed Cas9 DNA nucle- of crRNA and tracrRNA until a complex is formed. The ase is employed to recognize and cleave specific DNA crRNA and tracrRNA are purchased separately and then sequences after forming a ribonucleoprotein (RNP) assembled into a gRNA complex so that the crRNA can complex with a guide RNA (gRNA) [3]. This gRNA is a be re-designed for each new protospacer the user desires duplex that is composed of a CRISPR RNA (crRNA) to target. Next, purified Cas9 enzyme was mixed with the and a transactivating CRISPR RNA (tracrRNA) [3]. crRNA-tracrRNA complexes and incubated to allow for The crRNA contains a 20-base region designated as the the formation of Cas9 RNPs. Cas9 concentrations of 1 “protospacer”, which guides specific DNA cleavage by and 0.5 µg/µl were utilized based on optimization experi- binding to the complementary protospacer in the tar- ments in our laboratory. The Cas9 RNP complexes were get genome [4]. However, Cas9-mediated DNA cleavage then used for standard transformation of A. fumigatus occurs only if the protospacer is followed by an “NGG” protoplasts along with 2 μg of hygromycin resistance cas- protospacer adjacent motif (PAM) in the target genome sette that is flanked by 35 base pair homology regions as [5]. Several CRISPR/Cas9 systems have been developed a repair template. The major advantages of this in  vitro in Aspergillus species [6–11]. In most of these organisms, assembly system are the simplicity (i.e., does not require the Cas9 enzyme and gRNA are introduced via expres- strain construction) and the potentially portability from sion constructs that are either contained within autono- strain to strain. Our system generated nearly 100% gene ΔKu80 mously replicating plasmid or are integrated into the targeting in the ΔakuB mutant, increased gene dele- genome. Those that employ plasmids, control Cas9 activ - tion frequencies in the wild type strain Af293 from the ity through the presence or absence of selective agents typical ~ 5% up to ~ 74%, and produced gene targeting in the medium for plasmid maintenance whereas those efficiencies of ~ 90% in a clinical isolate [13]. Although designed for integration typically rely on regulatable pro- gene targeting was greatly improved by our method, and moters. CRISPR/Cas9 systems are becoming ever more is generally highly efficient in CRISPR/Cas9 methods sophisticated in Aspergillus, as evidenced by recent work developed for other the Aspergillus species, the potential showing that highly-efficient marker-free gene editing for off-target mutations has largely not been addressed. can be accomplished in A. nidulans, A. niger, and A. ory- Multiple recent studies have highlighted the potential zae strains defective in NHEJ [11]. In addition, highly effi - for CRISPR/Cas9-based gene editing systems to induce cient multi-site targeting is now possible in each of these unwanted off-target mutations [14–16]. This is typically Aspergillus species [11]. In A. fumigatus, the original believed to occur by promiscuous induction of dou- use of CRISPR/Cas9 involved strains that constitutively ble strand DNA breaks at sites other than the intended expressed Cas9 from an integrated construct [7]. A later protospacer. To ensure the reliability of our in  vitro iteration in A. fumigatus employed autonomously repli- assembly system, we sought to examine the impact of cating plasmids for Cas9 and gRNA expression and also exogenous Cas9 addition during the transformation pro- utilized selectable marker cassettes (“repair templates” in cess on the induction of genomic mutations in A. fumiga- CRISPR/Cas9 terminology) with microhomology regions tus using transformants from the studies we previously that are only 35–50  bp in length, showing that efficient performed in the Af293 genetic background. gene targeting can be accomplished in A. fumigatus with only small regions of DNA homology [12]. Although Results CRISPR/Cas9 gene targeting appears highly efficient To examine the potential for off-target mutations induced in A. fumigatus, the systems in place thus far rely on by the transient presence of gRNAs and the Cas9 nucle- genetically altering strains to express the required Cas9 ase in our system, we completed next-generation whole- genome sequencing on a subset of transformants isolated Al Abdallah et al. Fungal Biol Biotechnol (2018) 5:11 Page 3 of 8 from our previous study and performed a comparative allowed for colony color-based identification of homolo - analysis of the relative numbers and characteristics of gous integrations. The major difference between the sec - genomic mutations. In our previous study, reporting the ond and third experiments of our previous study was the optimization of in  vitro assembled Cas9 RNPs for gene use of either high (1 µg/µl) or low (0.5 µg/µl) concentra- targeting in A. fumigatus, we generated multiple transfor- tions of the Cas9 nuclease, respectively. Therefore, the mants from three basic experimental designs. In the first, experimental groups for our whole genome analyses in we utilized the wild type reference isolate, Af293, and the current study included three isolates from each of performed a transformation using standard protoplasting these three transformation conditions: no Cas9 (stand- protocols to ectopically integrate a hygromycin selection ard transformation protocol), 0.5  µg/µl Cas9 and 1  µg/ cassette, hygR [13]. In the second and third experiments, µl Cas9 (Table  1). As a reference, we also sequenced the we used Af293 to perform targeted CRISPR/Cas9-medi- parent strain, Af293. A. fumigatus Af293 is the genome ated integrations of the same hygR cassette to delete the reference isolate with a well-annotated genome. How- coding region of a polyketide synthase, pksP [13]. The ever, this parent isolate was re-sequenced to account pksP locus was chosen for protocol optimization as it for genomic mutations that may have arisen during the repetitive sub-culturing of this strain in our laboratory. Average genome coverage ranged from 38× to 62× for all isolates (Table 2). Table 1 Strains used in this study Comparative bioinformatics analyses revealed that Strain name Background multiple genomic mutations were present in the transfor- mants of all strains, regardless of the presence or absence Af293 Wild type of Cas9. In the experimental group lacking Cas9, a group NC1 hygR (no Cas9) average of 373 ± 28 mutations were identified (Table  2). NC2 hygR (no Cas9) Considered alone, this finding demonstrates the potential NC3 hygR (no Cas9) mutagenic nature of A. fumigatus protoplast transforma- HC4 ΔpksP‑hygR (1 µg/µl Cas9) tion. This standard transformation protocol may induce HC5 ΔpksP‑hygR (1 µg/µl Cas9) intense, albeit temporary, cellular stress as it requires the HC6 ΔpksP‑hygR (1 µg/µl Cas9) enzymatic digestion of the cell wall to release membrane- LC7 ΔpksP‑hygR (0.5 µg/µl Cas9) bound protoplasts followed by recovery on an osmoti- LC8 ΔpksP‑hygR (0.5 µg/µl Cas9) cally stabilized agar medium. Transformants from both LC9 ΔpksP‑hygR (0.5 µg/µl Cas9) the high (1 µg/µl) and low (0.5 µg/µl) Cas9 concentration For strains names, NC = no Cas9, HC = high Cas9, and LC = low Cas9. Indicated experiments displayed total numbers of genomic muta- at the right are the concentrations of Cas9 used to produce each mutant strain. hygR = hygromycin resistance cassette; pksP = polyketide synthase; ΔpksP- tions similar to the no Cas9 control, with an average of hygR = pksP locus replaced by hygR Table 2 Cas9-mediated gene deletion is not associated with increased genomic mutations in A. fumigatus NC1 NC2 NC3 HC4 HC5 HC6 LC7 LC8 LC9 Average coverage 53× 62× 60× 58× 54× 61× 56× 49× 38× Total mutations 396 342 385 345 326 307 345 366 474 Average total muta‑ 373 ± 28 SD 326 ± 19 SD (p > 0.05) 395 ± 69 SD (p > 0.05) tions Analysis based on type of mutation identified SNPs 371 318 363 321 314 292 326 347 439 Indels 25 24 22 24 12 15 19 19 35 Analysis based on location of mutation Intergenic 380 (96%) 331 (97%) 379 (98%) 331 (96%) 320 (98%) 301 (98%) 339 (98%) 359 (98%) 446 (94%) Average intergenic 363 ± 28 SD 317 ± 15 SD (p = 0.04) 381 ± 57 SD (p > 0.05) Coding region 16 (4%) 11 (3%) 6 (2%) 14 (4%) 6 (2%) 6 (2%) 6 (2%) 7 (2%) 28 (6%) Average coding 11 ± 5 SD 9 ± 5 SD (p > 0.05) 14 ± 12 SD (p > 0.05) region Displayed are the total and average number of mutations among the three experimental groups: no (“NC”—0 µg/µl), low (“LC”—0.5 µg/µl) and high (“HC”—1 µg/ µl) levels of Cas9. For the intergenic and coding region mutation rows, the numbers in parentheses represent the percent (%) of total mutations. For the average mutations per group, the mean ± standard deviation (SD) is provided. The Student’s t test assuming unequal variance was used for statistical comparisons of the Cas9 (HC or LC) and the no-Cas9 (NC) groups and p values are presented Al Abdallah et al. Fungal Biol Biotechnol (2018) 5:11 Page 4 of 8 326 ± 19 and 395 ± 69, respectively (Table  2). Among all total mutations) than those from the high concentration transformants, only a small subset of the identified muta - Cas9 experiment (~ 0.5% of total mutations). Thus, we tions were located within coding regions of the genome, interpret these mutations not as a consequence of Cas9 as the vast majority (> 96% for all isolates) were intergenic presence but as variability among isolates resulting from (Table 2). A single transformant (LC9) from the low Cas9 the standard protoplast transformation process. concentration experiment was found to contain relatively Cas9-induced double strand breaks (DSBs) can be higher levels of total mutations, with a total of 474, and repaired by two major pathways in the cell: the NHEJ a slightly increased distribution of these mutations into DNA repair pathway or homology directed repair [17]. coding regions (~ 5.9% of total compared to an average The NHEJ repair pathway is error prone and induces of ~ 2.6% for all other transformants). However, as this small insertion and deletion (indel) events into the transformant is an isolate from the low Cas9 concentra- genome [17]. Therefore, a negative consequence of Cas9- tion experiment, this apparent increase in total number mediated gene editing could be the promiscuous induc- of mutations and distribution towards coding regions is tion of DSBs resulting in increased indels throughout the not likely associated with the activity of Cas9. Further affected transformant(s). To see if this specific type of supporting this assertion, we found no statistically signif- off-target mutation may exist in our collection, we also icant difference in the total number of mutations identi - analyzed transformants from each experimental group fied among the experimental groups (Table  2). Therefore, for the disproportional generation of indels versus sin- our analyses revealed no Cas9 concentration-dependent gle nucleotide polymorphisms (SNPs). To do this, the increase in induction of genomic mutations. The muta - total genomic mutations (intergenic and coding region) tions identified in our study are likely induced by the were classified into either SNPs or indels and the result - transformation process, including cell wall digestion. ing numbers were averaged for each experimental group. To further examine if the presence of Cas9 during Our data indicated that the generation of SNPs was transformation may influence the type of mutation intro - favored over indels (1–80 nucleotides in length) within duced, we analyzed additional characteristics of the iden- each experimental group and no significant differences tified genomic variations. Those mutations identified in the relative amounts of either mutation were noted within coding regions displayed comparable distribu- (Fig. 1). When expressed as a percent of total mutations, tions among the categories defined in Table  3. In general, the non-Cas9 mediated transformation generated strains most mutations were located within the 3′ UTRs of genes containing 6.4% indels, whereas the high and low con- regardless of the presence or absence of Cas9. The only centration Cas9 transformants had 5.2 and 6.2% indels, discrepancy noted was that all transformants within the respectively. Therefore, the concentration of Cas9 is not high and low concentration Cas9 groups displayed a low associated with a disproportional increase in the num- number of non-synonymous mutations identified within ber of indels between experimental groups. Although coding regions, whereas the standard non-Cas9 transfor- our protospacers were designed to have minimal off-site mants were free of non-synonymous mutations (Table 3). complementarity, some level of similarity between our However, similar to our findings with total numbers of designed sequences and distant areas of the genome is mutations, the low concentration Cas9 transformants unavoidable. To ensure that off-target complementarity contained more non-synonymous mutations (~ 0.8% of of our gRNA complexes was not promiscuously driving Table 3 Cas9-mediated gene deletion does not cause alterations in the types of coding region mutations in A. fumigatus 0 µg/µl Cas9 1 µg/µl Cas9 0.5 µg/µl Cas9 NC1 NC2 NC3 HC4 HC5 HC6 LC7 LC8 LC9 3′ UTR 9 9 2 9 2 – 1 1 10 5′ UTR 3 – 1 2 – 1 1 – 1 Frameshift 1 – – 1 – – – – – Intron 2 2 2 1 1 1 2 3 2 Non‑synonymous – – – 1 1 3 2 1 6 Start lost – – – – 1 – – – – Synonymous – – – – 1 1 – 1 9 Splice region 1 – 1 – – – – 1 – Shown are the numbers and types of identified mutations located within coding regions of transformants within the three experimental groups Al Abdallah et al. Fungal Biol Biotechnol (2018) 5:11 Page 5 of 8 CRISPR/Cas9 components, this system can be utilized to study gene and pathway function across many isolates. Multiple studies have indicated that CRISPR/Cas9-based gene editing is potentially associated with off-target mutations whereas many studies have also described this system as highly specific [18]. Off-target mutations could be due to promiscuous activity of the Cas9 nucle- ase, yet more often seem to be caused by non-specific binding of gRNA [18]. Using the system we have adopted for A. fumigatus, both of these potential issues are prop- erly addressed [13]. The Cas9 nuclease is introduced as a purified protein and, therefore, is only transiently pre - sent. Compared to systems that constitutively or condi- tionally express Cas9, this should reduce the potential for unwanted mutagenesis of the genome generated by pro- miscuous Cas9 activity. In fact, in a few yeast and para- site studies, expression of CRISPR/Cas9 components was associated with toxicity [19–21]. Our system avoids this problem entirely. Additionally, using the rules for PAM selection, protospacer design, and RNP assembly that we have previously reported [13], we were able to generate Cas9-transformants that did not show increased genomic Fig. 1 Cas9‑mediated gene deletion does not cause a dispropor ‑ variability. Because the genomic variation identified in tional increase in SNPs or indels in A. fumigatus. Segregation of genomic mutations into SNPs (b) and indels (a) revealed that the our study is not correlated with CRISPR/Cas9 methods, concentration of Cas9 was not positively associated with an increase our data also suggest that either the single sub-culture in a specific subset of mutation. Shown are the average number of of a strain of A. fumigatus within the laboratory or the mutation events within each experimental group. Student’s t test stress induced by our standard protoplast transformation assuming unequal variance was utilized for statistical comparison procedure induces genomic variation. between the “no Cas9” and either the low (0.5 µg/µl) or high (1 µg/µl) Cas9 group Although whole genome sequencing of transformants has proven useful for the analysis of off-target effects in many systems, this technique does have the limitation of being unable to identify Cas9-induced DSBs that are per- Cas9 to generate unwanted DSBs and subsequent off-tar - fectly repaired. With this caveat in mind, whole genome get mutations, we finally interrogated each transformant sequencing has been successfully applied to characterize genome for variations surrounding ten potential off- the potential for off-target effects of Cas9-based transfor - target sites. These off-target sites were defined as areas mations in the plant pathogen Ustilago maydis [22]. This where the protospacer had twelve or more base pairs of study relied on expressing Cas9 from a strong constitutive complementarity. For both the 5′ and 3′ protospacer, our promoter on a self-replicating plasmid. After transforma- analysis found zero genomic mutations in these areas tion, strains could then be cured of this plasmid to avoid (data not shown). Together, these data indicate that continuous passage and growth in the presence of Cas9. CRISPR/Cas9 editing by our methods is highly specific in Whole genome sequencing of transformants acquired A. fumigatus. in this study revealed that none of the identified genome mutations were likely to be due to Cas9 activity mediated Discussion by gRNA binding [22]. Our results support the same con- The implementation of novel gene editing technolo - clusion when purified Cas9 is added exogenously to A. gies in A. fumigatus, like CRISPR/Cas9, is a critical step fumigatus. It is of note, however, that our study only inves- toward making significant advances in studies related to tigated two concentrations of Cas9, did not examine the virulence and antifungal drug resistance in this impor- effects of varying amounts of crRNA or tracrRNA or the tant human pathogen. We have shown that our system, ratios of Cas9 to tracrRNA and crRNA, and only inves- relying on in vitro assembled Cas9-RNP complexes, pro- tigated off-target effects upon targeting of only one gene. duces highly efficient gene targeting in multiple genetic It is possible that by targeting a different protospacer or backgrounds of A. fumigatus [13]. Because it does not by significantly altering the Cas9 RNP composition, a dif - rely on strains that have been genetically engineered to ferent outcome might have been observed. However, even increase homologous recombination rates or to express Al Abdallah et al. Fungal Biol Biotechnol (2018) 5:11 Page 6 of 8 though off-target effects might increase under these other resuspended in 800  μl of buffer AP1 and 8  μl of RNase conditions, further efforts that try to minimize them A and vigorously vortexed. Following 3  h of incubation can be pursed. For example, bioinformatic tools are now at 65  °C, the fungal lysate was centrifuged for 5  min at available to interrogate genomes for potential off-target 20,000×g and the supernatant was transferred to a new sites, including Cas-OFFinder and Cas-Designer [23, 1.5  ml tube. Next, 260  μl of buffer P3 were added to the 24]. Databases like these can aid in protospacer selection supernatant and the mixture was incubated for 5 min on and gRNA design to minimize the potential for off-target ice, then centrifuged for 5  min at 20,000×g. The super - mutations when targeting a new genomic locus. Also, natant was then transferred to QIAshredder spin column multiple studies into ways to minimize off-target muta - (700 μl at a time) and centrifuged for 2 min at 20,000×g. tions in CRISPR/Ca9 systems have been published and The flow-through was collected into a 2 ml tube without new techniques are constantly being pursued. One tech- disturbing the pellet, and 1.5 volumes of buffer AW1 was nique might be to limit the time of Cas9 activity in the cell added and mixed by pipetting. The mixture was trans - via use of photoactivatable split-Cas9 [25]. Our system ferred (700 μl at a time) into a DNeasy Mini spin column accomplishes limited Cas9 activity through introduction and the genomic DNA was allowed to bind to the col- of the Cas9 enzyme. However, if required, the specificity umn membrane by centrifuging for 1  min at ≥ 6000×g. of our system could also be further bolstered by employ- The genomic DNA was washed first with 700 μl of AW2 ing high-fidelity or rationally engineered Cas9 enzymes buffer and centrifuged for 1 min at ≥ 6000×g, followed by with increased specificity or through the use of truncated a second washing step using 300  μl of Buffer AW2 and versions of single-gRNAs [26–28]. centrifugation at 20,000×g for 5  min. The second wash - ing step was essential to remove any residual ethanol Conclusions on the membrane before elution step. The spin column The data provided here demonstrate that CRISPR/Cas9- was transferred to a new 1.5  ml tube. For the elution of mediated gene targeting, using our in  vitro assembled genomic DNA, 100  μl Buffer 5  mM Tris–HCl (pH 8.5) Cas9-RNP system, does not cause an increase in genomic was added to the center of the column and the column variation over standard transformation protocols. We was incubated at room temperature for 5  min, followed also identified no disproportional Cas9-dependent by a centrifugation step for 1  min at ≥ 6000×g. Final increase in SNPs or indels among treated strains. There - DNA concentrations were quantified using Nanodrop fore, Cas9-mediated gene deletion using in  vitro assem- and Qubit Fluorometer, following the manufacturer’s bled Cas9-RNPs coupled with microhomology repair protocol. templates is a reliable method for generating targeted mutations in A. fumigatus. Library preparation and bioinformatics analyses Library preparations and genome sequencing reactions Methods were performed at the University of Alabama at Birming- Strains and culture conditions ham Heflin Center for Genomic Science. The Qiagen Strains used for this study are listed in Table  1. Strain QIAseq FX DNA prep kit was used for library prepara- Af293 is the A. fumigatus reference genome isolate [29] tions, following the manufacturer’s instructions. Paired and all transformant strains employed for whole genome end 300 base pair sequencing reads were generated on sequencing were generated, as part of a previous study, the Illumina MiSeq following standard protocols. Bioin- in this genetic background [13]. All strains were main- formatics services were provided by code4DNA (www. tained on glucose minimal media (GMM) agar [30], sup- code4DNA.com). Reads from each sample were aligned plemented with hygromycin (150  µg/ml) for selection. using bwa mem (v0.7.15) to the A. fumigatus reference Conidia were harvested in water from three-day old genome build A_fumigatus_Af293_version_s03-m05-r05 plates and enumerated by hemocytometer. downloaded from AspGD.org [32]. Samtools (v1.3) fix - mate and rmdup were used to remove PCR duplicates Genomic DNA extraction [33]. Sequence mutations were called using FreeBayes Genomic DNA was extracted following a slight modifi - (v1.1.0) with the haploid population-based model [34]. cation of previous published protocols, using the Qia- Low quality (QUAL < 30) and low depth (DP < 10) muta- gen DNeasy Plant Mini Kit [31]. Briefly, strains were tions were filtered out using VCFtools (v0.1.15) which inoculated in GMM broth at a conidial density of 10 was also used to remove variant calls where no sequence conidia/ml and incubated for 20 h at 37 °C with shaking reads were available for at least one sample. The popu - at 250  rpm. Mycelia were harvested by vacuum filtra - lation.vcf file was split into individual samples using tion and 300 mg of a semi-dry mycelial mat was crushed VCFtools vcf-subset. Mutations were annotated using under liquid nitrogen. The resulting mycelial powder was snpEff (v4.3r) and VCFtools vcf-isec was used to select Al Abdallah et al. Fungal Biol Biotechnol (2018) 5:11 Page 7 of 8 10. Weyda I, Yang L, Vang J, Ahring BK, Lübeck M, Lübeck PS. A comparison of mutations found in each affected samples but not in the Agrobacterium‑mediated transformation and protoplast ‑mediated trans‑ Ref. [35]. formation with CRISPR–Cas9 and bipartite gene targeting substrates, as effective gene targeting tools for Aspergillus carbonarius. J Microbiol Methods. 2017;135:26–34. Abbreviations 11. Nødvig CS, Hoof JB, Kogle ME, Jarczynska ZD, Lehmbeck J, Klitgaard DK, CRISPR: clustered regularly interspaced short palindromic repeats; crRNA: Mortensen UH. Efficient oligo nucleotide mediated CRISPR–Cas9 gene CRISPR RNA; DSB: double strand break; GMM: glucose minimal media; gRNA: editing in Aspergilli. Fung Genet Biol. 2018. https://doi.org/10.1016/j. guide RNA; Indel: insertion/deletion; RNP: ribonucleoprotein; SNP: single‑ fgb.2018.01.004. nucleotide polymorphism; tracrRNA: trans‑activating crRNA. 12. Zhang C, Meng X, Wei X, Lu L. Highly efficient CRISPR mutagenesis by microhomology‑mediated end joining in Aspergillus fumigatus. Fung Authors’ contributions Genet Biol. 2016;86(Supplement C)):47–57. QAA performed the transformation of strains to obtain isolates for genome 13. Al Abdallah Q, Ge W, Fortwendel JR. A simple and universal system for sequencing and isolated genomic DNA for sequencing studies. WG cultured gene manipulation in Aspergillus fumigatus in vitro‑assembled Cas9‑ guide and prepared the selected strains for analysis. JRF contracted for the genomic RNA ribonucleoproteins coupled with microhomology repair templates. sequencing and bioinformatics analyses. QAA, WG, AMV, ACOS and JRF con‑ mSphere. 2017. https://doi.org/10.1128/msphere.00446‑17. tributed to data interpretation and manuscript preparation. All authors read 14. Zhang X‑H, Tee LY, Wang X ‑ G, Huang Q‑S, Yang S‑H. Off‑target effects in and approved the final manuscript. CRISPR/Cas9‑mediated genome engineering. Mol Ther Nucleic Acids. 2015;4:e264. https://doi.org/10.1038/mtna.2015.37. Acknowledgements 15. Schaefer KA, Wu W‑H, Colgan DF, Tsang SH, Bassuk AG, Mahajan VB. The authors would like to thank code4DNA (www.code4DNA.com) for con‑ Unexpected mutations after CRISPR–Cas9 editing in vivo. Nat Methods. tributing written summaries of how bioinformatics analyses were performed. 2017;14:547–8. 16. Zhang Q, Xing H‑L, Wang Z ‑P, Zhang H‑ Y, Yang F, Wang X‑ C, Chen Q‑ J. Competing interests Potential high‑frequency off‑target mutagenesis induced by CRISPR/Cas9 The authors declare that they have no competing interests. in Arabidopsis and its prevention. Plant Mol Biol. 2018;96:445–56. 17. Lieber MR. The mechanism of double‑strand DNA break repair by Ethics approval and consent to participate the nonhomologous DNA end joining pathway. Ann Rev Biochem. Not applicable. 2010;79:181–211. 18. O’Geen H, Yu AS, Segal DJ. How specific is CRISPR/Cas9 really? Cur Opin Funding Chem Biol. 2015;29:72–8. This work was supported by NIH Grant R01AI106925 to J.R.F. 19. Ryan OW, Skerker JM, Maurer MJ, Li X, Tsai JC, Poddar S, Lee ME, DeLoache W, Dueber JE, Arkin AP, Cate JHD. Selection of chromosomal DNA librar‑ ies using a multiplex CRISPR system. eLife. 2014;3:e03703. https://doi. Publisher’s Note org/10.7554/elife.03703. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 20. Generoso WC, Gottardi M, Oreb M, Boles E. Simplified CRISPR–Cas lished maps and institutional affiliations. genome editing for Saccharomyces cerevisiae. J Microbiol Methods. 2016;127:203–5. Received: 2 March 2018 Accepted: 18 April 2018 21. Jiang W, Brueggeman AJ, Horken KM, Plucinak TM, Weeks DP. Successful transient expression of Cas9 and single guide RNA genes in Chla- mydomonas reinhardtii. Eukaryot Cell. 2014;13:1465–9. 22. Schuster M, Schweizer G, Reissmann S, Kahmann R. Genome editing in Ustilago maydis using the CRISPR–Cas system. Fungal Genet Biol. 2016;89:3–9. References 23. Park J, Bae S, Kim J‑S. Cas‑Designer: a web ‑based tool for choice of 1. Krappmann S, Sasse C, Braus GH. Gene targeting in Aspergillus fumigatus CRISPR–Cas9 target sites. Bioinformatics. 2015;31:4014–6. by homologous recombination is facilitated in a nonhomologous end‑ 24. Bae S, Park J, Kim J‑S. Cas‑ OFFinder: a fast and versatile algorithm that joining‑ deficient genetic background. Eukaryot Cell. 2006;5:212–5. searches for potential off‑target sites of Cas9 RNA‑ guided endonucleases. 2. da Silva Ferreira ME, Kress MRVZ, Savoldi M, Goldman MHS, Härtl A, Bioinformatics. 2014;30:1473–5. Heinekamp T, Brakhage AA, Goldman GH. The akuBKU80 mutant defi‑ 25. Nihongaki Y, Kawano F, Nakajima T, Sato M. Photoactivatable CRISPR– cient for nonhomologous end joining is a powerful tool for analyzing Cas9 for optogenetic genome editing. Nat Biotechnol. 2015;33:755–60. pathogenicity in Aspergillus fumigatus. Eukaryot Cell. 2006;5:207–11. 26. Kleinstiver BP, Pattanayak V, Prew MS, Tsai SQ, Nguyen NT, Zheng Z, Joung 3. Sander JD, Joung JK. CRISPR–Cas systems for editing, regulating and JK. High‑fidelity CRISPR–Cas9 nucleases with no detectable genome ‑ targeting genomes. Nat Biotechnol. 2014;32:347–55. wide off‑target effects. Nature. 2016;529:490–5. 4. Shah SA, Erdmann S, Mojica FJM, Garrett RA. Protospacer recognition 27. Slaymaker IM, Gao L, Zetsche B, Scott DA, Yan WX, Zhang F. Ration‑ motifs. RNA Biol. 2013;10:891–9. ally engineered Cas9 nucleases with improved specificity. Science. 5. Mojica FJM, Díez‑ Villaseñor C, García‑Martínez J, Almendros C. Short motif 2016;351:84–8. sequences determine the targets of the prokaryotic CRISPR defence 28. Fu Y, Sander JD, Reyon D, Cascio VM, Joung JK. Improving CRISPR–Cas system. Microbiology. 2009;155:733–40. nuclease specificity using truncated guide RNAs. Nat Biotechnol. 6. Nødvig CS, Nielsen JB, Kogle ME, Mortensen UH. A CRISPR–Cas9 2014;32:279–84. system for genetic engineering of filamentous fungi. PLoS ONE. 29. Nierman WC, Pain A, Anderson MJ, Wortman JR, Kim HS, Arroyo J, Berri‑ 2015;10:e0133085. https://doi.org/10.1371/journal.pone.0133085. man M, Abe K, Archer DB, Bermejo C, Bennett J, Bowyer P, Chen D, Collins 7. Fuller KK, Chen S, Loros JJ, Dunlap JC. Development of the CRISPR/Cas9 M, Coulsen R, Davies R, Dyer PS, Farman M, Fedorova N, Fedorova N, system for targeted gene disruption in Aspergillus fumigatus. Eukaryot Feldblyum TV, Fischer R, Fosker N, Fraser A, García JL, García MJ, Goble A, Cell. 2015;14:1073–80. Goldman GH, Gomi K, Griffith‑ Jones S, Gwilliam R, Haas B, Haas H, Harris 8. Katayama T, Tanaka Y, Okabe T, Nakamura H, Fujii W, Kitamoto K, Maruy‑ D, Horiuchi H, Huang J, Humphray S, Jiménez J, Keller N, Khouri H, Kita‑ ama J. Development of a genome editing technique using the CRISPR/ moto K, Kobayashi T, Konzack S, Kulkarni R, Kumagai T, Lafton A, Latgé J‑P, Cas9 system in the industrial filamentous fungus Aspergillus oryzae. Li W, Lord A, Lu C, Majoros WH, May GS, Miller BL, Mohamoud Y, Molina Biotechnol Lett. 2016;38:637–42. M, Monod M, Mouyna I, Mulligan S, Murphy L, O’Neil S, Paulsen I, Peñalva 9. Zhang C, Meng X, Wei X, Lu L. Highly efficient CRISPR mutagenesis by MA, Pertea M, Price C, Pritchard BL, Quail MA, Rabbinowitsch E, Rawlins microhomology‑mediated end joining in Aspergillus fumigatus. Fung N, Rajandream M‑A, Reichard U, Renauld H, Robson GD, de Córdoba Genet Biol. 2016;86:47–57. SR, Rodríguez‑Peña JM, Ronning CM, Rutter S, Salzberg SL, Sanchez M, Al Abdallah et al. Fungal Biol Biotechnol (2018) 5:11 Page 8 of 8 Sánchez‑Ferrero JC, Saunders D, Seeger K, Squares R, Squares S, Takeuchi 32. Li H, Durbin R. Fast and accurate short read alignment with Burrows– M, Tekaia F, Turner G, de Aldana CRV, Weidman J, White O, Woodward J, Yu Wheeler transform. Bioinformatics. 2009;25:1754–60. J‑H, Fraser C, Galagan JE, Asai K, Machida M, Hall N, Barrell B, Denning DW. 33. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abe‑ Genomic sequence of the pathogenic and allergenic filamentous fungus casis G, Durbin R. The sequence alignment/map format and SAMtools. Aspergillus fumigatus. Nature. 2005;438:1151–6. Bioinformatics. 2009;25:2078–9. 30. Shimizu K, Keller NP. Genetic involvement of a cAMP‑ dependent protein 34. Garison E, Marth G. Haplotype‑based variant detection from short ‑read kinase in a G protein signaling pathway regulating morphological and sequencing. 2012. arXiv preprint: arXiv:1207.3907[q‑bio.GN]. chemical transitions in Aspergillus nidulans. Genetics. 2001;157:591–600. 35. Cingolani P, Platts A, Wang LL, Coon M, Nguyen T, Wang L, Land SJ, Lu X, 31. Hagiwara D, Takahashi H, Watanabe A, Takahashi‑Nakaguchi A, Kawamoto Ruden DM. A program for annotating and predicting the effects of single S, Kamei K, Gonoi T. Whole‑ genome comparison of Aspergillus fumigatus nucleotide polymorphisms, SnpE. F ff ly. 2012;6:80–92. strains serially isolated from patients with Aspergillosis. J Clin Microbiol. 2014;52:4202–9. Ready to submit your research ? Choose BMC and benefit from: fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Fungal Biology and BiotechnologySpringer Journals

Published: Jun 5, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off