Introduction

Bioeconomy has emerged as a novel economic and sustainability paradigm to replace the use of nonremovable resources by promoting feedstock conversion into biofuels and several bioproducts. Feedstock processing in biorefineries, has driven the search for new sources of enzymes and more efficient production strategies1,2,3.Among fungi, Trichoderma reesei RUT C-30 is the most used since it hyper secretes biomass-degrading enzymes for several industrial applications, but also is well known for their long genetic improving history4. Another cellulase producers include species from Penicillium, Aspergillus, and Trichoderma5. However, bioprospecting for novel fungal species and secretomes as sources of single enzymes or cellulase cocktails is needed6,7. In that sense, A. fumigatus, are also being investigated for their biotechnological potential for enzyme prodution8. A. fumigatus stands out as producer of cellulases, xylanases, and ligninases, for processing lignocellulosic compounds from plant materials9,10,11, making it an attractive candidate for enzyme production. A noteworthy case is A. fumigatus LMB-35Aa, an alkaline cellulase producing saprophytic strain isolated from Peruvian rainforest soil, which is considered safe due to the absence of certain virulence-related genes found in clinical strains12,13. However, identification and characterization of these kind of cellulases in this strain has not been deeply investigated.

Several strategies have been developed to improve production yield and catalytic efficiency of cellulolytic fungal enzymes. For instance, surface adhesion fermentation (SAF)14, such as those mediated by biofilms, have shown to enhance productivity in A. niger ATCC10864 strain15. Additionally, conventional genetic manipulation, including random mutagenesis, have been successfully employed to obtain hypersecreting mutants such as Trichoderma reesei RUT-C304 as well as Aspergillus species16,17. Site-directed mutagenesis has also considered for catalytic enhancement of cellulolytic enzymes18,19,20.

Also, heterologous production of modified enzymes is usually required, making it a challenge to produce the complete set of cellulases. Heterologous production of cellulases in bacterial or yeast host could be limited because of incorrect folding, N- and O-glycosylation, and inefficient secretion21,22. Hence, the development of an optimized heterologous expression system for complete lignocellulose degradation remains a costly and labor-intensive process.

Alternatively, the insertion of cellulases into the fungal genomes has been explored22,23,24. However, this approach is limited by random gene integration or by the low homologous recombination rate mediated by long arms in the insertion cassette for gene disruption. Consequently, the low selection rate of disruptive events hinders the identification of a sufficient number of transformants through homologous integration of a gene cassette25,26. Therefore, the development of new strategies to meet the growing demand for cellulolytic enzymes in various industries is crucial.

In this context, precise methods for introducing genetic modifications in fungi are required to improve cellulase production through genetic engineering. The CRISPR/Cas9 gene editing system has emerged as a highly efficient tool for targeted genetic recombination of filamentous fungi27,28,29, including non-models strains or those that have not been molecularly characterized30,31. Despite the homologous direct recombination (HDR) or non-homologous end-joining (NHEJ) repair pathways, an alternative independent mechanism, termed Microhomology-Mediated End Joining (MMEJ), has been already described in A. fumigatus31. Indeed, gene editing studies in A. fumigatus have primarily focused on genes related to pathogenicity using clinical and commercial isolates such as Af293 and CEA1032. The CRISPR/Cas9 system relies on the use of a complex, formed by an endonuclease (Cas9) and a single guide RNA molecule (sgRNA), which directs the complex to its target sequence and allows the endonuclease to cut at a specific point in the fungal genome33. The high efficiency of CRISPR/Cas9 in genetic recombination, and its potential to generate mutant strains both through gene knock-out and by knock-in make it a valuable tool for optimizing enzymatic production in non-model filamentous fungi27,28,29,34. A. fumigatus possess a predominant NHEJ repair system, making HDR less efficient. Knockout strains of the aku80 gene have been developed to address this, achieving gene deletion rates of 46 to 74% using 35–50 bp homology regions, compared to low gene targeting efficiencies (5–20%) with classical genetic manipulation methods35.

CRISPR/Cas technology has already been used for heterologous expression of different cellulases in Saccharomyces cerevisiae36,37, as well as for improvement of cellulase production38,39,40,41,42,43 or to produce various secondary metabolites in filamentous fungi44,45. Although heterologous expression could facilitate the obtention of enzymatic cocktails by genomic design, currently, there are no reports of gene editing for heterologous cellulase production in Aspergillus and related genera. Therefore, in this study the CRISPR/Cas9 system was used to achieve the heterologous production of endoglucanase EglA from the fungus A. niger ATCC 10,864 in the A. fumigatus LMB-35Aa wild-type strain. EglA is an endoglucanase enzyme, known for its activity at an optimally pH of  546, which aligns with the optimal pH range for A. fumigatus LMB-35Aa under the conditions used in this study13. To facilitate the selection process of knock-in events through the identification of albino colonies, eglA gene was integrated through the MMEJ pathway into the conidial melanin pksP gene locus.

Results

Construction of CRISPR/Cas9 vector and gene cassettes for targeted mutagenesis of A. fumigatus

For this study, the pksP gene was chosen as target because it is responsible for conidial pigmentation and induces albino colony formation upon its loss of function31,47. The strategy used to efficiently deliver both the sgRNA and the Cas9 protein to the pksP target locus closely followed the approach used by Nødvig30. For this purpose, the pJB112 vector was designed, which incorporates cas9 and pksP gRNA sequences, regulated by the Aspergillus nidulans tef1 and gpdA promoters, respectively. To generate the gRNA cassette, pksP gRNA was fused with the Hammerhead (HH) and hepatitis delta virus (HDV) ribozymes from the pFC334 vector and subsequently ligated it into the pFC332 vector (see Fig. 1 for details).

Figure 1
figure 1

Overview of the construction of the CRISPR/Cas9 vector to edit A. fumigatus LMB-35Aa genome. Adapted from Nødvig et al. (2015)30. AMA1: sequence for plasmid replication. HH and HDV: Ribozyme sequences. Intron: intronic region of the gpdA gene. Dark blue represents a region which is complement to the dark blue found in the yA crRNA. ampR: antibiotic resistance gene to ampicillin. hygR: antibiotic resistance gene to hygromycin. ori: replication origin. Ptef1 and PgpdA: promoter regions. Ttef1 and TtrpC: terminator regions. BglII and PacI: restriction sites. cas9: gene for the Cas9 endonuclease. yA crRNA: CRISPR RNA sequence (corresponds to the protospacer sequence) of the yA gene. Dark blue represents a region which is complement to the dark blue found in the intronic region. gRNA: guide RNA sequence, contains the crRNA of the pksP gene and tracrRNA regions. BglII-Fw and Rv-PacI: primers containing the BglII and PacI restriction site sequences. gRNA-Fw and Rv-gRNA: primers containing the protospacer sequence of the pksP target gene (green line).

Full size image

Integration cassettes of gfp and eglA genes (~ 2000 bp) were generated by fusion PCR and contained the trpC promoter and terminator flanked by 39 nt at its 5´and 3´ends that are homologous to the adjacent ends of the cut site.

Inactivation of pks P gene in the native A. fumigatus LMB-35Aa strain by CRISPR/Cas9 targeted mutagenesis

To validate the effectiveness of gene edition into pksP target site, a first attempt to transform A. fumigatus LMB-35Aa protoplasts only with the pJB112 vector was done. At about 55 (10.22%) of the regenerated colonies exhibited an albino phenotype (Fig. 2) and 9 of these were selected for further verification of targeted mutagenesis. Sequencing of PCR fragments covering the genomic region adjacent to the Cas9 cleavage site revealed four different mutations (Fig. 3). Mutation 1 consisted of a large insertion of 88 nucleotides, generating a stop codon near the cutting site. Interestingly, the insertion exhibited by this mutant corresponded to the AMA1 region of the pFC332 plasmid. The three remaining mutations consisted of small indels: mutant 2 showed a cytosine deletion, mutant 3 a Thymine insertion, and mutant 4 a deletion of 5 nucleotides. All of them have a reading frame shift generating a stop codon near the cutting site, which explain pksP loss of function.

Figure 2
figure 2

Targeted mutagenesis of pksP gene of A. fumigatus LMB-35Aa mediated by CRISPR/Cas9. Regenerated colonies of A. fumigatus LMB-35Aa transformed with pJB112 plasmid. Red arrows indicate albino colonies.

Full size image
Figure 3
figure 3

CRISPR/Cas9 mutations targeted to pksP locus of A. fumigatus LMB-35Aa. DNA sequence alignments of A. fumigatus Af293 wild-type strain and PCR fragments neighboring the Cas9 cleavage pksP site amplified from A. fumigatus LMB-35Aa wild-type and edited mutant strains. Protospacer (pksP target site) and PAM sequences are shown in green and red, respectively. Insertions are shown in blue, and hyphens indicate deleted nucleotides.

Full size image

It should be mentioned that no sequences matching the pksP protospacer-PAM were identified within the plasmid pFC332, which harbor the AMA1 sequence, that could potentially be cleaved by the sgRNA-Cas9 rib onucleoprotein complex.

Knock-in of exogenous genes into pks P gene locus of A. fumigatus LMB-35Aa by CRISPR/Cas9 targeted mutagenesis

To obtain knock-in mutants, two separated co-transformations of A. fumigatus LMB-35Aa wild-type strain were carried out with vector pJB112 and each integration cassettes (gfp or eglA). The first one carried the green fluorescent protein (GFP) gene as reporter to validate the edition, while the second one contained an A. niger endoglucanase gene (eglA) aimed at enhancing the endoglucanase activity of A. fumigatus LMB-35Aa.

Albino colonies were evaluated under a phase-contrast microscope showing typical conidiophore and hypha structures (Fig. 4a). Also, fluorescence microscopy revealed GFP expression in twelve co-transformed albino colonies with a fluorescence pattern distributed throughout the hypha and conidiophores, except in the vacuoles (Fig. 4b). Subsequently, to confirm the insertion of the gfp gene at the pksP target site (Fig. 5), these mutants were assessed by PCR using primers that flank the Cas9 cleavage site as shown in (Fig. 5). Among the twelve evaluated clones, only one of them (LMB072) amplified the fragment of expected size (~ 2000 bp), which contains gfp cassette integrated at the Cas9 cleavage site; the remaining clones showed amplicon sizes of around 500 bp, similar to the wild-type strain (Supplementary Fig. S1). However, the integration of gfp cassette in the genomes of these edited strains was also confirmed by PCR (~ 800 bp) (Supplementary Fig. S1).

Figure 4
figure 4

A. fumigatus LMB-35Aa albino knock-in mutant expressing GFP protein. (a) Phase-contrast and (b) fluorescence micrographs showing conidiophores and conidiospores at magnification of 400X.

Full size image
Figure 5
figure 5

Construction and insertion of gfp and eglA gene cassettes into the pksP locus of A. fumigatus LMB-35Aa genome. (a) Genomic localization of the insertion site of the gene cassettes PtrpC-gfp-TtrpC or PtrpC-eglA-TtrpC in the A. fumigatus LMB-35Aa genome, the Cas9 cleavage site in the pksP gene and the microhomologous regions (39 bp) of the gene cassette, and the Protospacer Adjacent Motif (PAM) region “CGG” of the gRNA (in red) are observed. (b) Comparative representation of the genomic region in an edited mutant, displaying the integrated fragment of the gfp or eglA gene, in contrast to the wild-type strain. Additionally, the depiction includes the position of primers used for verification and the amplification size of ~ 2000 bp in the presence of the genetic cassette and ~ 500 bp in the wild-type strain. pksP Fw: pksP-Seq-F, pksP Rv: pksP-Seq-R, Fw: (PtrpC)-gfp-F or (PtrpC)-AniEglA-F, Fw: (TtrpC)-gfp-R or (TtrpC)-AniEglA-R.

Full size image

For eglA cassette, after A. fumigatus LMB-35Aa co-transformation, thirteen albino colonies were isolated and evaluated by PCR by employing primers that flank the Cas9 cleavage site as shown in (Fig. 5). The fragment of the expected size (~ 2000 bp), which contains eglA cassette integrated at the Cas9 cleavage site, was amplified in only three of the clones evaluated; however, two of these clones also exhibited several non-specific bands. As observed in the case of gfp cassette integration, amplicons showing sizes like the wild-type strain (~ 500 bp) were also obtained. Likewise, the integration of eglA gene in the genomes of these edited strains was also confirmed by amplifying the Open Reading Frame (ORF) of this gene by PCR and the clone which exhibited only bands of the expected sizes in both amplifications (LMB073) was selected for further biochemical assays (Supplementary Fig. S2).

Additionally, to identify putative off-target sites that would explain the nonspecific integration of gfp or eglA genes, a comprehensive in silico analysis was performed. The results revealed the absence of off-target sites both 1 kb upstream and downstream of the target A. fumigatus LMB-35Aa pksP sequence. Nevertheless, a sequence (ACGCTATTCCCGCGG) exhibiting the protospacer-PAM match was identified 100 kb downstream of the target sequence, although with one mismatch (in bold).

Effect of the eglA gene expression on A. fumigatus LMB-35Aa endoglucanase activity

Once the successful integration of eglA cassette in knock in edited mutant strains was confirmed (Supplementary Fig. S2), eglA gene expression was validated by using RT-PCR of A. fumigatus LMB073 mutant strain (Supplementary Fig. S3). Subsequently, this selected knock in strain was cultivated in a biofilm system, using carboxymethyl cellulose (CMC) as a substrate to induce cellulase production. Notably, the specific endoglucanase activity exhibited by the LMB073 strain (94.89 ± 8.1 U/g) was ~ 40% higher (26.89 U/g) than the wild-type strain (68 ± 4.6 U/g) (Student’s t test, p-value < 0.05).

Discussion

Bioeconomy scenario has emerged for sustainable production of biofuels and biproducts from renewable resources, mainly lignocellulose, stressing the need to obtain novel enzymes or enzymatic cocktail sources, that included cellulases and other biomass degrading enzymes48. Complementarly different approaches has been used for strain improvement to produce cellulases, that include genetic and metabolic engineering, mutagenesis, omics and genomic engineering, and CRISPR system49,50.

Currently, the CRISPR/Cas9-mediated mutagenesis has revolutionized genetic manipulation through the development of versatile methods to precisely achieve gene disruptions and gene replacements in a wide range of filamentous fungi28,29,30,34. Despite some limitations such as incomplete gene annotation of the target strain, potential off-target effects, and/or low efficiency of detection and cleavage of the sgRNA, the application of the CRISPR/Cas9 mutagenesis system still constitutes a very promising tool for precise and efficient gene editing in non-model fungi29,30,32, with valuable applications in diverse fields such as the development of antifungal agents, production of natural compounds, enzymes, and optimization of fermentation processes28,29,45,51.

Although successful applications of CRISPR/Cas9 gene editing have been achieved in fungi, there are currently limited reports on its use for cellulase heterologous expression or cellulase production enhancement29,32,39,40,41,43,51,52. In this study, we aimed to address this gap by using the CRISPR/Cas9 system to edit A. fumigatus LMB-35Aa strain and inserting the exogenous eglA endoglucanase gene. This genetic modification was combined with a biofilm fermentation system, which is considered suitable for enhancing cellulase production in A. niger15, to promote higher cellulase production in the edited mutants.

While biofilm formation is commonly associated with infection processes and fungal pathogenicity53, it has also been described in the case of opportunistic A. fumigatus clinical strains in immunosuppressed patients53,54,55. However, an expression analysis of pathogenicity-related genes of the non-clinical strain used in this study revealed that virulence is influenced by multiple factors beyond biofilm formation and growth temperature13. This underscores the complexity of fungal pathogenicity and the specific characteristics of the non-clinical strain used in this study.

With the aim of editing the A. fumigatus LMB-35Aa genome, the vector pJB112 with the AMA1 sequence, which has been shown to be an efficient replication origin in many fungal species56,57, and the hygromycin resistance gene (hygR) as a selection marker, was designed and constructed . The pksP gene involved in the coloring of the A. fumigatus conidia, was selected as the target gene, which facilitated the phenotypic identification of albino edited mutants31,47 and allowed a successful gene edition with a ~10 % efficiency. Previous work in A. fumigatus has reported gene editing efficiencies from 25 to 53 % using the same protospacer (target) sequence31,47; while in other gene editing attempts in diverse Aspergillus species, targeting genes such as wA, yA and pyrG, the efficiencies have varied widely (10–100%)30,58,59,60. However, it is worth noting that the 100% editing efficiency reported for the yA gene was calculated based on only the three obtained transformants58. Furthermore, Nødvig et al. (2018) improved it editing efficiency by adding long homology arms61. Additionally, different gene editing efficiencies (10-44 %) have been reported in Trichoderma reesei39,41,51, Penicillium oxalicum41, Talaromyces pinohpilus40 and Rasamsonia emersonii43 by knocking-out cellulase transcriptional regulators such as ace1 and cre1, as well as by introducing additional copies of the xyr1 regulator gene with the aim of enhancing cellulase production. Apparently, the ploidy level is another crucial factor that could also affect editing efficiency, being lower in multinuclear fungi30,58,59,62.

Notably, the ~40 % increase of endoglucanase activity obtained with the eglA knock-in edited mutant LMB073 (Fig. 6) surpasses the improvements reported by other authors using CRISPR/Cas9 technology. For instance, Singh et al.43 achieved a 22 % increase in R. emersonii by knocking-out the ace1 gene43; while Zhang et al.51 only achieved between 4 and 7 % increase by knocking-in a xyr1 gene into the ace1 locus in T. reesei51. Our findings indicate that even though previous studies used regulatory genes to improve cellulase production, the increase in activity was not as significant as observed in this study. Therefore, this study represents the first report of successful improvement of endoglucanase activity in Aspergillus by knocking-in an exogenous gene (eglA) through CRISPR/Cas9 mediated targeted mutagenesis, highlighting its potential as a source of cellulases or enzymatic cocktails by genomic design, and setting a new benchmark for further improvement of LMB-35Aa strain.

Figure 6
figure 6

Endoglucanase specific activity of A. fumigatus LMB-35Aa eglA knock-in edited strain vs. wild-type strain. Endoglucanase activity was determined using CMC as a substrate from biofilm culture supernatants obtained at 72 h of growth at 28 °C. The activity was measured in 50 mM acetate buffer at pH 4.8, using 1% CMC as the substrate, at 50 °C, and incubated for 30 min.LMB-35Aa: A. fumigatus LMB-35Aa wild-type strain; LMB073: A. fumigatus LMB-35Aa eglA knock-in edited strain. Specific activity (U/g) = Endoglucanase activity (U/L)/soluble secreted protein (g/L). Data represents the mean of three replicates. Error bars represent the standard deviations. The statistical difference between both strains was determined by Student’s t-test p < 0.05 indicated by *.

Full size image

Although apparently gene cassettes could be integrated into an alternative genomic locus of A. fumigatus, this could be due to a non-specific integration of the expression cassette via NHEJ, but also because of potential off-target effects. In fact, off-target effects constitute a potential limitation associated with CRISPR/Cas9 technology, particularly in eukaryotic organisms. However, no putative off-target sites were found in the regions 1 kb up- and downstream of the target pksP sequence in A. fumigatus LMB-35Aa genome, but a sequence of 15 nucleotides exhibiting only one mismatch with the protospacer-PAM sequence was identified 100 kb downstream. Whether this region could have been the nonspecific insertion site will need to be confirmed by PCR and further sequencing.

On the other hand, regarding the insertion of 88 nucleotides detected in one of the knock-out edited strains (mutant 1), which corresponded to the AMA1 region in pFC332 plasmid, it could be attributed to the cleavage mediated by the sgRNA-Cas9 ribonucleoprotein complex. Nonetheless, not founded sequences matching the pksP protospacer-PAM within the pFC332 sequence, strongly suggests that the 88 bps AMA1 sequence inserted in that mutant was not due to an additional RNA-guided Cas9 activity at these loci on pFC332 plasmid.

It is also well known that gene manipulation in filamentous fungi is constrained by the low efficiency of homologous recombination in DNA-mediated transformation and the high frequency of ectopic integrations of the transforming DNA molecule63. Often, at least 1 kb of homologous DNA arm is required to achieve homologous recombination efficiencies of about 10%26,64. In contrast, using CRISPR/Cas9 technology, Zhang et al.31 reported an editing accuracy of 95-100 % using very short homologous arms (35 bp) in A. fumigatus. Furthermore, the authors suggest that the microhomology-mediated repair system differs from NHEJ, based on their experiments with the wild-type strain and the knock-out of the ku80 gene (a key gene in the non-homologous end-joining repair system). Since it is the only one report mentioning this hypothesis, we cannot rule out the possibility that, due to the short homologous arms used in this study (~39 bp), the gene cassette may have been inserted at non-specific sites.

Finally, based on the notable improvement of endoglucanase activity obtained in the eglA knock-in edited A. fumigatus mutant, it would be expected that integrating the advantages of biofilm fermentation systems with the use of improved fungal strains could have a significant biotechnology impact, leading to higher yields, productivity and ultimately, profitability. However, future enhancements will focus on optimizing the LMB-35Aa strain by integrating additional genetic modifications to further boost production of enzymes or enzymatic cocktails by genomic design.

Methods

Strains and culture media

Fungal strains A. fumigatus LMB-35Aa12 and A. niger ATCC 10,864 were maintained on potato dextrose agar (PDA) plates at 28 °C. Fungal biomass was obtained by vacuum filtration using a Whatman No. 1 filter paper (Whatman International Ltd, Kent, UK) from potato dextrose broth (PDB) cultures or, alternatively, from cellulase production medium (KH2PO4, 2 g/L; CaCl2.2H2O, 0.3 g/L; MgSO4.7H2O, 0.3 g/L; urea, 0.3 g/L; (NH4)2SO4, 1.4 g/L; peptone, 1 g/L; Tween 80, 2 mL/L; FeSO4.7H2O, 5 mg/L; MnSO4.2H2O, 1.6 mg/L; ZnSO4.7H2O, 1.4 mg/L; CoCl2.6H2O, 2 mg/L)65. The bacterial strain E. coli DH5α was kindly donated by Dr. Richard Garrat (IFSC, USP) and was used for storage and construction of all vectors. The details of the microbial strains used are presented in (Supplementary Table S1).

Biofilm fermentation system (BF)

To obtain the spore inoculum for biofilm formation, A. fumigatus LMB-35Aa strain was grown in PDA for 3 days at 28 °C. After complete sporulation, the spores were resuspended in 10 mL of a 0.1% (v/v) Tween 80 solution and finally adjusted to a concentration of 1 × 105 spores/mL. The formation of biofilms was conducted on polyester supports following the procedure outlined by Gamarra et al. (2010)15, with some modifications. Briefly, a pre-weighted piece of polyester cloth (3.1 cm × 3.1 cm) was placed in 250 mL flasks containing 70 mL of sterile dH2O. Subsequently, flasks were inoculated with 3% (v/v) of the previously prepared spore suspension (inoculum) and incubated at 28 °C and 175 rpm for 30 min to allow spore adhesion to the supports. Then, dH2O was carefully decanted, and the non-adhered spores were removed by washing the cloths twice with the same volume of sterile dH2O for 15 min under the same incubation conditions. Washed cloths were then transferred to new 250 mL flasks containing 70 mL of cellulase production medium using 0.5% carboxymethylcellulose (CMC) as the only carbon source and incubated at the same conditions for 72 h. The cultures were performed in triplicate. Finally, culture supernatants were collected and stored at 4 °C for subsequent analysis of enzyme activity and quantification of soluble protein by using the Pierce™ BCA Protein Assay kit (ThermoFisher), while the fungal biofilms were washed three times with cold sterile dH2O and stored at −80 °C until nucleic acid extraction.

Nucleic acid extraction and cDNA synthesis

Plasmid DNA was extracted using the ZR Plasmid Miniprep kit (Zymo Research) according to the manufacturer’s instructions. Genomic DNA extraction from fungal biomass/biofilms was carried out following the protocol described by Cenis66, with some modifications.

Total RNA extraction from Aspergillus strains was performed from ~ 50 mg of previously pulverized biomass grinded with liquid nitrogen, using the Direct-zol RNA Miniprep kit (Zymo Research) following the manufacturer's instructions.

Nucleic acid samples were eluted in nuclease-free water and stored at −20 °C (DNA) or −80 °C (RNA). Sample concentrations were determined by spectrophotometry using a Nanodrop 2000 (Thermo Scientific) and the integrity of DNA/RNA samples was evaluated by agarose gel electrophoresis.

cDNA synthesis was carried out from 2 μg of total RNA in a final volume of 25 μL using a reverse transcription mix containing 200 U of M-MLV RT enzyme (Promega), 25 U of RNAse inhibitor (RNasin, Promega), 0.5 μM of dNTPs mix, 1X RT buffer, and 1 μg of Oligo(dT)18 as a primer. Reactions were carried out at 42 °C for 60 min and stored at −20 °C until required.

Construction of CRISPR/Cas9 vector and gene cassettes

All PCRs for the construction of the vector and gene cassettes were performed using Phusion DNA Polymerase (Thermo Fisher), according to the manufacturer’s specifications. For gene cassette constructions, DNA fragments were amplified separately and then fused using the fusion PCR technique according to Shevchuk (2004)67. The list of vectors and primers used are presented in Supplementary Table S2 and Table S3, respectively.

To construct the editing vector, the strategy outlined by Nødvig et al. (2015) was closely followed. The similarity of the protospacer sequence to be used (5´CTCAGCGCACGCTCTTCCCG 3´; according to Fuller et al.47) with the pksP gene of A. fumigatus LMB-35Aa strain (Fig. 3) was confirmed. Subsequently, primers encompassing the aforementioned protospacer sequence from the pksP target gene were designed (pksP-gRNA-pFC334-R and pksP-gRNA-FC334-F). Simultaneously, primers that flanked the promoter and terminator regions of the gRNA cassette were designed, containing the BglII and PacI restriction sites (pFC334-BglII-F and pFC334-PacI-R). These primers were employed to amplify two distinct fragments from plasmid pFC334 (Addgene plasmid #87846). Resulting amplicons, both carrying the pksP gRNA protospacer sequence, were annealed and fused by PCR using pFC334-BglII-F and pFC334-PacI-R primers, thus forming the pksP gRNA cassette which was subsequently ligated into the plasmid pFC332 (Addgene plasmid #87845), both previously digested with PacI and BglII. The resulting vector containing the gRNA for knocking out pksP gene, designated as pJB112, was transformed into E. coli DH5α competent cells by heat shock68. Positive clones were verified by colony PCR using the conditions established by the GoTaq enzyme protocol (Promega). For more details of the strategy followed for vector construction, see (Fig. 1).

To construct the integration cassettes encoding the green fluorescent protein (GFP) or the EglA endoglucanase, the following steps were followed: i) for the GFP cassette, on the one hand, the trpC promoter and terminator were amplified from pSilent1 plasmid69 using (pksP1)-PtrpC-F/(gfp)-PtrpC-R and (gfp)-TtrpC-F/(pksP1)-TtrpC-R primers, respectively; on the other hand, reporter gfp gene was amplified from the pERA-4 plasmid70 using (PtrpC)-gfp-F and (TtrpC)-gfp-R primers. Resulting amplicons were annealed and fused by PCR using primers (pksP1)-PtrpC-F and (pksP1)-TtrpC-R, thus forming the GFP cassette; ii) for the eglA cassette, three amplicons corresponding to promoter trpC, eglA gene and trpC terminator were fused. For this purpose, PtrpC and TtrpC were amplified from pSilent1 plasmid69 using (pksP1)-PtrpC-F/(AniEglA)-PtrpC-R and (AniEglA)-TtrpC-F/(pksP1)-TtrpC-R primers, respectively; eglA gene was amplified from A. niger ATCC 10864 cDNA using (PtrpC)-AniEglA-F and (TtrpC)-AniEglA-R primers. Resulting amplicons were annealed and joined by PCR using primers (pksP1)-PtrpC-F and (pksP1)-TtrpC-R, thus forming the eglA cassette. Both replacement gene cassettes contained 39 bp microhomologous arms at the 5ʹ ends of the promoter and 3ʹ ends of the terminator flanking the pksP gene.

Off-target analysis

The chop-chop tool (https://chopchop.cbu.uib.no/) was used for off-target analysis based on the genome of A. fumigatus Af293 strain and, for the LMB-35Aa strain, alignments between the protospacer and PAM sequences against the genome of the LMB-35Aa strain (MCQI00000000.2) were conducted using the BLAST tool (https://blast.ncbi.nlm.nih.gov).

Protoplast formation

For fungal protoplast formation, the procedure outlined by Yelton et al.71 was followed with some modifications. Briefly, 3-day-old spores from A. fumigatus LMB-35Aa strain grown on PDA medium were harvested using 0.1 % (v/v) Tween 80. Then, a suspension of 5 x 108 spores was inoculated into 250 mL flasks containing complete medium (CM: 0.5 % malt extract, 0.5 % yeast extract, and 0.5 % glucose) and incubated at 30 °C and 175 rpm for 14 hours. Young mycelia were recovered in 50 mL tubes by filtration through Mira-Cloth (22–25 µm pore size) and washed with 0.6 M MgSO4·7H2O. Washed fungal biomass (3 g) was transferred to a 250 mL flask containing 25 mL of digestion buffer (1.2 M MgSO4·7H2O in 10 mM sodium phosphate buffer at pH 5.8). Mycelial cell wall was digested using 300 µg of lysing enzyme (Sigma, L1412), for this purpose, the mixture was incubated on ice for 5 min then, BSA was added (3 mg per gram of biomass) and incubated for 4 hours at 30 °C and 100 rpm. Subsequently, formed protoplasts were harvested by filtration through Mira-Cloth into 50 mL tubes and precipitated by the addition of STC buffer (1.2 M sorbitol, 10 mM CaCl2, 10 mM Tris-HCl pH 7.5) in a 1:1 (v/v) ratio and centrifuged at 6000 rpm for 10 min. Finally, protoplasts were resuspended in 1 mL of STC buffer and 1 mL of 40 % PEG 3350 was added for storage at −80 °C until further use.

Polyethylene glycol (PEG)-mediated protoplast transformation

For fungal transformation, the procedure outlined by Yelton et al. (1984)71 was followed with some modifications. Briefly, 1 x 106 fungal protoplasts resuspended in 200 μL of STC buffer were incubated with ~5 µg of the corresponding DNA vector in a 15 mL tube for 20 minutes on ice. Then, 1.25 mL of PEG buffer (PEG 3350 60%, 50 mM CaCl2, 50 mM Tris-HCl pH 7.5) was added and incubated for 20 minutes at room temperature. Subsequently, 5 mL of STC buffer was added and gently homogenized. Transformed protoplasts were centrifuged at 4000 rpm for 10 minutes at 4 °C and resuspended in 1 mL of STC buffer. Finally, protoplasts were plated on Stabilizing Minimal Medium (SMM) following the formulation outlined by Shimizu & Keller (2001)72, with some modifications. It is noteworthy that our medium diverged from Shimizu & Keller´s formulation by the exclusion of MgSO4, adjustment of FeSO4.7H2O and (NH4)6Mo7O24.4H2O to 0.16 and 0.11 g/L, respectively. To solidify the medium, agar was added to a final concentration of 1.6 %. Then, an overlay of 0.9 % agar containing 300 µg/mL hygromycin was added and once solidified, incubated at 30 °C. The appearance of regenerated protoplasts was evaluated at 24, 48, and 72 hours. Finally, transformed colonies were verified in SMM medium containing hygromycin 300 µg/mL. pskP knock-out colonies were verified by their phenotype (albino colonies) and by further PCR and amplicon sequencing using pksP-Seq F and pksP-Seq R primers.

For knock-in gene editing, 2 µg of eglA or gfp gene cassette DNA was added to the transformation reaction. Positives clones were verified by PCR, using primers that flank the Cas9 cut site (pksP-Seq F and pksP-Seq R) for the insertion into the pksP gene locus. Finally, the clone expressing eglA cassette were confirmed by measuring its endoglucanase activity and by gene expression analysis by RT-PCR.

Endoglucanase activity assay

ß-endoglucanase activity was quantitatively determined in 96-well microplates according to the methodology described by Xiao et al. (2005)73 with certain modifications. Briefly, 30 µL of culture supernatant (enzyme) and 60 µL of 1 % CMC (substrate) prepared in 50 mM acetate buffer pH 4.8 were added into each microwell, and incubated for 30 min at 50 °C. The detection and quantification of the reducing sugars produced during CMC hydrolysis were carried out according to Miller´s methodology74 using DNS (3,5-dinitrosalicylic acid) reagent. For this, 90 µL of DNS reagent was added and incubated at 95 °C for 5 min, followed by cooling to 12 °C for 10 min. Finally, 100 µL of the colored mixture was transferred to a 96-well microplate and the absorbance was measured at 540 nm using an EPOCH 2 microplate spectrophotometer (BioTek). An enzyme blank and substrate blank were also included in each assay. The concentration of glucose released by the secreted enzymes was determined by interpolating from a standard curve constructed with known concentrations of glucose. One enzyme unit (U) was defined as the amount of enzyme required to release 1 µmol of reducing sugar equivalents per minute under the defined assay conditions. Specific activity was expressed in enzymatic units per milligram of proteins (U/mg).

Fluorescence and phase-contrast microscopy

To verify gene editing of gfp knock-in strains, a confocal laser scanning microscope (OLYMPUS BX61) equipped with DAPI, WBV, GFP, TRITC, and DICT filters, and a Nikon UPlanSApo VC 40X objective (NA = 0.95) was used. For this purpose, fungal spores were mounted on a glass slide and the GFP protein was excited/monitored at 458/488 nm. Digital images were acquired by using FLUOVIEW FV1200 software (FV10-ASW).

Statistical analysis

For the statistical analysis, Google Colab platform (https://colab.google/) was used. First, the normality of the data distributions for LMB-35Aa and LMB073 was assessed using the Shapiro–Wilk test, which confirmed normality for both groups. Then, Levene’s test was applied to verify the homogeneity of variances. Following these tests, a two-sample t-test (Student’s t-test) was performed.