Generation of Trichoderma harzianum with pyr4 auxotrophic marker by using the CRISPR/Cas9 system

Trichoderma harzianum is a filamentous fungus used as a biological control agent for agricultural pests. Genes of this microorganism have been studied, and their applications are patented for use in biofungicides and plant breeding strategies. Gene editing technologies would be of great importance for genetic characterization of this species, but have not yet been reported. This work describes mutants obtained with an auxotrophic marker in this species using the CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats)/ Cas (CRISPR-associated) system. For this, sequences for a guide RNA and Cas9 overexpression were inserted via biolistics, and the sequencing approach confirmed deletions and insertions at the pyr4 gene. Phenotypic characterization demonstrated a reduction in the growth of mutants in the absence of uridine, as well as resistance to 5-fluorotic acid. In addition, the gene disruption did not reduce mycoparasitc activity against phytopathogens. Thus, target disruption of the pyr4 gene in T. harzianum using the CRISPR/Cas9 system was demonstrated, and it was also shown that endogenous expression of the system did not interfere with the biological control activity of pathogens. This work is the first report of CRISPR Cas9-based editing in this biocontrol species, and the mutants expressing Cas9 have potential for the generation of useful technologies in agricultural biotechnology.

www.nature.com/scientificreports/ harzianum on plants are possible due to their ability to colonize and penetrate the roots of plants and to carry out symbiotic relationships 2,3,11,12 . Due the fact that T. harzianum is among the bioagents most used in today's agriculture worldwide 11,30,31 , there is increasing interest in understanding the modes of action of this biocontrol fungus and the underlying molecular processes in greater detail. The recent development of the CRISPR/Cas9 gene editing technique could form the basis for large-scale genetic manipulations of this biocontrol fungus, but the establishment of additional selection markers is also crucial. Thus far, only a limited number of selection markers have been available for genetic transformation of T. harzianum, and OMP-decarboxylase deletion (pyr −) has proved to be a reliable auxotrophic marker for filamentous fungi 5,14,22,32,33 . Furthermore, the effects of gene deletion together with Cas9 overexpression in a biocontrol fungus is innovative. The use of the CRISPR/Cas9 gene editing system to disrupt the pyr4 gene in T. harzianum represents a promising strategy for validating the technique in this fungus; it also prepares the ground for further work on gene editing and the functional analysis of this system during mycoparasitism.

Results and discussion
Since genetic tools have scarcely been developed for most filamentous fungus, it is currently difficult to employ genetic engineering in understanding the biology of Trichoderma spp. and to fully exploit them industrially 8,34 . Moreover, the frequency of homologous recombination in some species is traditionally very low, time-consuming and sometimes troublesome 16,19,20,33 . For these reasons, there is a demand for developing versatile methods that can be used to genetically manipulate this biocontrol species. Therefore, gene editing technologies represent a highly promising alternative in genetic engineering of T. harzianum and have prompted us to establish new mutant lines for large-scale genetic manipulations. To facilitate this, we have developed a CRISPR/Cas9-based system adapted for use in this biocontrol fungus.
To CRISPR/Cas9-mediated genome editing, both the endonuclease and the sgRNA need to be present in the nucleus of the target organism 13 . In order to create vectors suitable for pyr4 gene editing in T. harzianum, the respective Cas9 sequence was inserted in pNOM102 plasmid 35 , under control of the A. nidulans gpdA promoter and trpC terminator. Subsequently, the gRNA sequence for pyr4 was inserted in pLHhph1-tef1 36 plasmid, containing a hygromycin phosphotransferase gene (hyg) from E. coli as a dominant selectable marker. The resulting plasmids, pCas and pGpyr4 (Fig. 1), were used for fungal transformation procedure.
Protoplast transformation of Trichoderma species includes PEG/CaCl 2 , electroporation, and A. tumefaciensmediated strategies. However, preparation of protoplasts using various cell-wall degrading enzymes is timeconsuming and expensive. In this way, biolistic bombardment is simple and versatile, as plasmids can be delivered into Trichoderma intact conidia.
Disruption of pyr4 confers 5-fluoroorotic acid (5-FOA) resistance to T. harzianum. Mutants which are defective in pyr4 are prototrophic strains resistant to 5-FOA, which is converted by orotidine-5′-monophosphate (OMP)-decarboxylase to the toxic intermediate 5-fluoro-UMP 37 . In this work, biolistic have been employed successfully for introducing Cas9 and gRNA in T. harzianum, and positive transformants presented both plasmids by PCR, as described in the materials and methods section. Colonies began to appear 3 days after plating of conidia on selective medium containing 5-FOA and uridine. Fourteen transformants were generated in two bombardment experiments (12 plates), with an efficiency of between 0 and 3 co-transformants per plate. Four T. harzianum mutants (named ΔP3, ΔP4, ΔP7 and ΔP13) that showed the codon-optimized Cas9 gene (Fig. 1B) after selection in 5-FOA medium and single spore isolation (Fig. 1C) were used for assays.
Correspondingly, sequencing approaches were used to carry out a comparative pyr4 analysis with the wildtype strain, and indels at the gRNA target were shown for all mutants (Fig. 1D). Thus, the CRISPR/Cas9 technique enabled the production of T. harzianum strains with an auxotrophic marker that also expressed the Cas9 gene. From a practical perspective, our work introduces a powerful genome-editing approach in mitotically stable mutants with endogenous pyr4 gene disruption accomplished by Cas9 expression. This system could be versatile and simple, as new mutagenesis can be achieved in T. harzianum lines by re-transforming with a single plasmid containing RNA guide.
One of the most important advancements in recent years, for improving the performance of research with Trichoderma species is the development of auxotrophic strains. Disruption of pyr4 also generates auxotrophic strains defective for uridine (uracil). In our work, we present the successful establishment of this selection marker for the genetic transformation of the biocontrol fungus T. harzianum. Indeed, results from assays in PDA medium without uridine demonstrated that mutants (ΔP3, ΔP4, ΔP7 and ΔP13) presented lower growth ratios compared to the wild-type (WT) strain ( Fig. 2A). Moreover, assays in PDA with uridine revealed that all mutants showed higher growth ratios compared to WT ( Fig. 2A). In relation to assays conducted in MEX medium, it was demonstrated that pyr4 disruption also reduced the mutants' growth ratio in the absence of uridine. However, addition of uridine to MEX medium only reestablished mutants' growth ratio similarly to WT (Fig. 2B). In this way, the CRISPR/Cas9 system caused insertions and deletions (indels) in target regions of the pyr4 gene and successfully interrupted gene function. In addition, phenotypic analyses confirmed that these mutants need a complement of uridine in the medium to present similar growth to the wild-type, thus certifying the presence of this convenient selection marker.
Microorganisms with pyr-negative marker are widely used in biotechnological processes, industries and research 14,21,25 . Indeed, genomic research with the industrial species T. reesei entered a new era after pyr-negative strains became available. Nevertheless, T. reesei with this auxotrophic marker has been produced using both traditional genetic approaches 22,25 and, recently, CRISPR/Cas9 technology 20 . Additionally, research using the CRISPR/Cas9 system reveals that these markers are exclusive to T. reesei 15,16,20,38 . These studies used protoplast or Agrobacterium transformation methods and described only in vitro transcription of gRNA 15 www.nature.com/scientificreports/ was successfully carried out by biolistic direct transformation of T. harzianum with the Cas9/gRNA complex, and it may be an alternative means to achieve fast gene disruption, while the overexpression of a codon-optimized Cas9 provides a means to speed up genome editing in this biocontrol fungus. In addition, the use of Cas9 and gRNA in separate plasmids allows the generation of new edition vectors by manipulating only the gRNA vector in a simple and cheaper manner. Mutants overexpressing Cas9 could be re-transformed with new gRNA vectors, taking advantage of the pyr4 auxotrophic marker. Despite these advantages, there has been no report of using such a technique in other Trichoderma biocontrol species. Trichoderma harzianum is a cosmopolitan filamentous fungus that displays a remarkable range of applications in agricultural biotechnology 2,3 . Because of its ability to antagonize plant-pathogens as well as stimulating plant growth and defense responses, some strains are used in bioformulation for biological control 2,11,26,27 . In this way, gene disruption has been a critical technique for improvement of T. harzianum strains and biocontrol studies.
The effects of pyr4 disruption and Cas9 overexpression on the mycoparasitic interaction between T. harzianum and fungal hosts were assessed in plate confrontation assays. In relation to S. sclerotiorum assays, we observed that the absence of uridine did not affect mutants' ability to mycoparasitize this pathogen, compared to the WT strain (Fig. 3A). However, confrontation assays carried out in the presence of uridine demonstrated that mutants decreased S. sclerotiorum overgrowth compared to wild-type (Fig. 3A).
Confrontation assays were also performed to compare mycoparasitic abilities of T. harzianum strains against F. oxysporum either in absence or presence of uridine (Fig. 3B). No differences between the tested strains were observed for the inhibition of F. oxysporum in all media analyzed (Fig. 3B). Bioassays with pathogens demonstrated that pyr4 gene disruption (OMP-decarboxylase), important for the pyrimidine synthesis pathway, in addition to Cas9 expression, did not reduce the mycoparasitic activity of mutants. Thus, our results underscore the use of the CRISPR/Cas9 system in T. harzianum has many prospects for functional analysis of biocontrol genes, metabolic modifications, and the selection or production of new strains for biotechnological uses.

Conclusion
For the first time, this work successfully established a promising approach for genome editing in the biocontrol fungus T. harzianum. Mutants produced with an auxotrophic marker and Cas9 overexpression provide a tool for functional analysis of biocontrol genes, selection of strains for bioformulations, and the generation of new strains for biotechnological uses.

Materials and methods
Microorganisms and culture conditions. Trichoderma harzianum ALL42 (Enzymology group collection-UFG/ICB) was used for this study. Fusarium oxysporum and Sclerotinia sclerotiorum were from the EMBRAPA-CNPAF culture collection. The microorganisms were maintained on potato/dextrose/agar (PDA) plates with periodic sampling and stored at 4 °C in EMBRAPA/CNPAF before use.

Preparation of microparticles and cells for bombardment, and biolistic co-transformation of T. harzianum.
Transformation procedure was based on previous protocols 18,39,40 , with some essential modifications described in the following. DNA was bound to 0.2-μ-diameter tungsten particles (M5, Sylvania Inc.) by mixing sequentially in a microcentrifuge tube: 50 μl microparticles (60 mg ml −1 in 50% glycerol), 5 μl (1 μg μl −1 ) of each plasmid constructed (pCas and pGpyr4), 50 μl CaCl 2 (2.5 M) and 20 μl spermidine free-base (100 mM). After 10 min incubation, the DNA-coated microparticles were centrifuged (15,000×g, 10 s) and the supernatant removed. The pellet was washed with 150 μl 70% ethanol and then with absolute ethanol. The final pellet was resuspended in 24 μl of absolute ethanol and sonicated for 2 s, just before use. Aliquots of 3 μl were spread onto carrier membranes (Kapton, 2 mil, DuPont) which were allowed to evaporate in a desiccator at 12% relative humidity.
The target material for transformation by microparticle bombardment was Trichoderma harzianum (ALL42) intact conidia. A suspension of conidia, previously produced by cultivation of the fungus on potato-dextrose agar was prepared by harvesting the conidia from the plate, suspending them in 0.9 M NaCl, and separating them from mycelial carryover by filtration through a column filled with glasswool. A conidial suspension (30 μl) containing 1.7 × 10 7 spores ml −1 was bombarded with the DNA-coated microparticles utilizing a high pressure helium-driven particle acceleration device built in our laboratory 40 . The relative humidity in the biolistic laboratory was 50%, the gap distance from shock wave generator to the carrier membrane was 8 mm, the carrier membrane flying distance to the stopping screen was 13 ram, the DNA-coated microparticles flying distance to the target was 80 mm, the vacuum in the chamber was 27 inches of Hg and the helium pressure utilized in all experiments was 1 200 psi. After the bombardment, transformants were incubated at 28 °C on yeast extract/agar (MEX) plates containing 5-FOA (1.5 g/L; Fermentas, St. Leon-Rot, Germany) and uridine (10 mM).
Selection and stabilization of co-transformants. Inoculated plates were incubated at 28 °C for up to 10 days during which plates were periodically examined directly for Trichoderma harzianum conidia development. Colonies appearing after incubation were picked using a sterile needle and transferred to fresh selective medium. Mutants were sub-cultured for a further three cycles of mycelial growth and conidiation. Molecular analysis of T. harzianum mutants and sequencing. Following three rounds of singlespore isolation, we obtained 14 mutants by phenotypic analysis (5′FOA resistance). Genomic DNA from four co-transformed strains were isolated as described previously 22 and screened by PCR amplification with primers specific for pCas cassette (Cas9_RNAgCheC: 5′-CTG CAA GGC GAT TAA GTT GG-3′/ Cas9_3897F: 5′-ACA GCA TAA GCA CTA CCT CG-3′) and also pGpyr4 vector (hygF:5′-CAC GTT GCA AGA CCT GCC TGAA-3′/ hygR:5′-TCC GGA TGC CTC CGC TCG AAGTA-3′). The amplification conditions were: an initial denaturation step of 2 min at 94 °C, followed by 30 cycles of 30 s at 94 °C, 30 s at 55 °C and 60 s at 72 °C, and a final extension step of 10 min at 72 °C. The pyr4 gene fragment, which was used for sequencing and further analysis, was amplified (pyrF: 5′-AGC TCT AAC CTG TGC CTG A-3′/ pyrR: 5′-AAG GTA GAG GAG CTC CCG -3′), cloned into the pGEMT-Easy vector according to standard procedures and sequenced using SP6 universal primer. DNA from the wild type (WT) strain was included as control.
Growth and direct confrontation assays. To analyze Trichoderma harzianum mutants for uridine auxotrophy, mycelium-covered plugs were placed at the center of fresh PDA or MEX plates supplemented with 10 mM uridine and incubated at 28 °C for 7 days. Antagonism activity of T. harzianum WT and mutants against pathogens was performed as a plate confrontation assay as described previously 41 , and colony diameter measurement was taken for a period of 7 days. Two pathogens (Sclerotinia sclerotiorum and Fusarium oxysporum) were