CRISPR/Cas9 system-directed gene mutagenesis without donor DNA in Myceliophthora thermophila

Because efficient expression of sgRNA requires a suitable promoter, a search for the native promoter of putative U6 snRNA genes in the genome of M. thermophila ATCC 42464 was conducted using bioinformatics tools. The putative RNA polymerase III U6 snRNA genes (U6) were located (http://genome.jgi.doe.gov) and a region approximately 500 bp upstream of the U6 gene was chosen as a trial promoter for expression of sgRNA (Fig. 1).

To investigate whether the constructed CRISPR/Cas9 system (Fig. 1) could perform gene editing in M. thermophila, a rapid testing system was designed based on amdS/fluoroacetamide (FAA) selection. This system was chosen because M. thermophila WT strains do not grow on acetamide as a nitrogen source and are resistant to certain FAAs. The transformants carrying the amdS gene were sensitive on FAA medium; thus, colonies harboring inactivated amdS could be isolated on FAA medium, thereby facilitating counter selection. Therefore, the amdS gene of A. nidulans was first artificially introduced into the M. thermophila WT genome through Agrobacterium-mediated transformation [12]. The transformant carrying the functional amdS gene was named M1. An sgRNA sequence containing a 20-nucleotide target (Table 1) of the amdS gene was designed (U6p-amdS-sgRNA) to perform the amdS gene editing. The transient PCR product of U6p-amdS-sgRNA was co-transformed with a Cas9 expression cassette into protoplasts of the recipient strain M1 (containing the amdS gene and FAA-sensitive) (Fig. 3a). Seven FAA-resistant colonies were obtained and their amdS gene mutations were verified by sequencing. The seven mutants displayed frame-shift mutations upstream of the PAM site (Fig. 3b) involving either a single-nucleotide insertion/deletion or large insertions of 30, 65, or 104 nucleotides (the 104-nucleotide insertion mutant was named the M2 stain, and was used as a host strain for subsequent engineering). In a control experiment in which transformation was conducted using the Cas9 plasmid only, no colonies were observed on FAA-containing medium. Taken together, these data demonstrated that the CRISPR/Cas9 system could mutate the target gene in the M. thermophila genome via NHEJ-mediated events during DSB repair.

Open in a separate window

Fig. 3

Introduction of directed mutations into the amdS gene in M. thermophila by sgRNA-guided Cas9. a Schematic illustration of amdS mutagenesis by the CRISPR/Cas9 system. The M1 strain carrying the amdS gene was co-transformed with U6p-amdS-sgRNA and Cas9 cassettes, and then positive transformants were selected on medium plates containing 2 mg/mL FAA. b Sequence alignment of the amdS target locus from FAA-resistant transformants. The wild-type target sequence of amdS is framed in blue. Red letters depict the protospacer adjacent motif (PAM)

CRISPR/Cas9 system-stimulated gene editing with donor DNA in Myceliophthora thermophila and M. heterothallica

CRISPR/Cas9 system-mediated HDR has been shown to work efficiently in many organisms, including several mesophilic fungi [33–40]. To determine whether the CRISPR/Cas9 system could also operate via HDR on target gene loci in M. thermophila, a transient single-sgRNA PCR cassette was transformed into ATCC 42464 WT protoplasts together with a Cas9 expression cassette and a donor DNA PCR product composed of the G418-resistance cassette PtrpC-neo and 5′ and 3′ homologous arms.

As a proof of principle, the carbon catabolite repression (CCR) transcription factor cre-1 [44] was chosen as a target locus for editing (Fig. 4a). The protospacer sequence targeted to cre-1 is shown in Table 1. On the basis of genomic PCR analysis using specific primer sets (Fig. 4a), the selected marker cassettes were confirmed to be correctly inserted into the desired locus (ΔMtcre-1) in 19 out of 20 randomly picked transformants (Fig. 4b). The homologous recombination (HR) efficiencies for all tested transformants are summarized in Table 2. Compared with transformation with donor DNA alone (20%), the HR frequency of gene replacement was as high as 95% after co-transformation using the CRISPR-Cas9 system and HR donor DNA (Table 2; Fig. 4b; Additional file 2: Fig. S2). The HR rates were low (15%) when Cas9 or sgRNA alone was used for the transformation (Table 2; Additional file 2: Fig. S2).

Open in a separate window

Fig. 4

Verification of cre-1 deletion in selected transformants. a Schematic of homologous recombination (HR) of the target gene cre-1 mediated by Cas9, gRNA and donor DNA. (b-c) PCR analysis of selected transformants of M. thermophila b and M. heterothallica c with one primer (cre1-out-F) located upstream of the 5′ flanking region of the genomic DNA and the other (cre1-in-R) located in the 3′ flanking region of the genomic DNA. The expected length of disrupted transformants was 1.9 kb, while that of the host strain, used as a negative control, was 1.0 kb (rightmost lane)

To evaluate the possibility of broader application of the above-described CRISPR/Cas9 system to other thermophilic fungi, it was used to perform genome editing in M. heterothallica, also with cre-1 as the target gene. The Cas9, U6p-cre1-sgRNA, and donor-cre1 constructs were co-transformed into M. heterothallica strain CBS203. Using both G418 and phosphinothricin selection, 20 transformants were chosen randomly and validated by PCR. As illustrated in Table 2, Fig. 4c, this system functioned well in CBS203.75, with the G418 marker cassette successfully integrated into the cre-1 locus of 18 transformants with 90% HR frequency. In the control experiments, only 15–20% disruption efficiencies were obtained when Cas9 or sgRNA was used for the transformation (Table 2; Additional file 2: Fig. S2).

Deletion of cre-1 led to increased cellulase production

Similar to the morphological alterations observed in cre1 deletion mutants of T. reesei, Neurospora crassa, and P. oxalicum [44–49], the M. thermophila ΔMtcre-1 mutant exhibited slower, denser growth on sucrose than the WT ATCC 42464 (Fig. 5a). The effect of the cre-1 deletion on cellulase and hemicellulase production was also investigated. Compared with the WT strain, the ΔMtcre-1 mutant secreted 3.7-fold higher amounts of extracellular protein into the Avicel medium after 4 days of cultivation (Fig. 5b, c). Xylanase, endoglucanase, and exoglucanase activities in ΔMtcre-1 were all increased approximately threefold compared with the WT strain in 2% Avicel (Fig. 5d–f). Similar to the phenotype of ΔMtcre-1, the M. heterothallica ΔMhcre-1 mutant also formed smaller, denser colonies relative to the WT strain CBS203 (Fig. 5a). When grown on medium containing 2% Avicel as the sole carbon source, the ΔMhcre-1 mutant displayed a pronounced increase in secreted protein levels and lignocellulase activities (Fig. 5b–f) compared with the WT strain CBS203 (Tukey’s HSD, p < 0.001).

Open in a separate window

Fig. 5

Phenotypic analysis of Δcre-1 and wild-type strains. a Colonies of Δcre-1 (ΔMtcre-1 and ΔMhcre-1) and wild-type strains (MtWT and MhWT) of M. thermophila and M. heterothallica on minimal medium plates after 2 days. b Sodium dodecylsulfate-polyacrylamide gel electrophoresis of secreted protein of Δcre-1 and wild-type strains after 4 days culture in 2% Avicel medium. c–f Assays for protein concentration and xylanase, endoglucanase, and exoglucanase activities of Δcre-1 and wild-type strains in 2% Avicel inducing medium after 3 and 4 days culture. Bars marked by asterisks in each group differ significantly from unmarked ones (Tukey’s HSD, p < 0.001). Error bars represent SD from three replicates

CRISPR/Cas9-mediated simultaneous multiplex-locus genome editing in Myceliophthora thermophila

Because the CRISPR/Cas9 system functioned very efficiently for single-gene (e.g., cre-1 locus) editing in M. thermophila, multiple gene disruption through a single transformation was investigated. In addition to cre-1, two genes involved in cellulase production were chosen as targets: an endoplasmic reticulum stress regulator res-1 homolog [50] and the gh1-1 gene encoding β-glucosidase [10, 49, 51–53]. M2, a strain with constitutive Cas9 expression and inactive amdS (through a large insertion of 104 nucleotides) that was derived from M1 (a strain carrying the amdS gene under the WT background, see Fig. 3), was used as the recipient strain in further experiments. First, three combinations of two-gene editing were performed (cre1 + res1, cre1 + gh1-1, and res1 + gh1-1; Table 2). The sgRNA and donor DNA of each of the two genes were co-transformed in a one-step transformation. A total of 23 transformants were randomly picked on selective plates containing G418 and phosphinothricin. All 23 randomly chosen transformants were then checked for both genes by diagnostic PCR. Using the CRISPR/Cas9 editing system, the efficiency of simultaneous disruption of the two genes was approximately 61–70% (Additional file 3: Fig. S3; Table 2). No obvious preference for a particular tested locus was observed in regard to disruption efficiency. A total of 10 Δcre1Δgh1-1, 11 Δcre1Δres1, and 11 Δgh1-1Δres1 homokaryotic double deletions were obtained (Additional file 3: Fig. S3).

Next, simultaneous deletion of all three loci (cre1, res1, and gh1-1) was conducted. Using the sgRNA and donor DNA sequences described above, all three sgRNA/donor DNA gene disruption sets were co-transformed together in the same molar amounts into the M2 strain. After 3 days of culture in selective medium, 23 transformants were randomly picked and subjected to PCR analysis. Seven transformants (homokaryon + heterokaryon) showed homologous recombination in the cre1, res1, and gh1-1 gene loci, and the efficiency of simultaneous triple recombination was about 30% (Additional file 4: Fig. S4; Table 2). Via one transformation, in particular, five of seven possible combinations of mutant genotypes were obtained, including the single-gene mutant Δcre-1, all three double-mutant genotypes (Δcre1Δgh1-1, Δcre1Δres1, and Δgh1-1Δres1) and the triple mutant Δcre1Δgh1-1Δres1, with only two single-gene mutants (Δres1 and Δgh1-1) missing (Additional file 4: Fig. S4).

Finally, simultaneous four gene disruption mediated by the CRISPR/Cas9 system was performed. In addition to cre1, res1, and gh1-1, the gene alp-1 [9], encoding alkaline protease, was chosen as the fourth deletion target. The experiment was performed as described above for three-gene deletion. Among 23 selected transformants, five clones (homokaryon + heterokaryon, 22%) were identified with all four genes disrupted, including three with homokaryotic quadruple deletions (ΔcΔgΔrΔa) (Additional file 5: Fig. S5). The copy number of the integrated neo marker in these transformants was determined by real-time quantitative PCR (RT-qPCR). Generally speaking, the copy number of the integrated marker was in good accordance with the number of target genes, which indicated that additional random integration of the marker did not happen often in this fungus (Additional file 6: Fig. S6).

Construction of Cas9 and sgRNA expression plasmids

All primer sequences used in this study are listed in Additional file 7: Table S1. All PCR products were amplified using Phusion high-fidelity DNA polymerase (Thermo Fisher, Waltham, MA, USA). A codon-optimized Cas9 gene with attached hac-1 (MYCTH_2310995) nuclear localization signals (NLS-Cas9-NLS) was synthesized by Life Technologies (Invitrogen) for expression in M. thermophila and M. heterothallica. The synthetic NLS-Cas9-NLS, the strong constitutive tef1 (MYCTH_2298136) promoter of M. thermophila, and the TtprC terminator were amplified using paired primers (Additional file 7: Table S1). With the aid of a NEB Gibson assembly kit, these amplification products were assembled to form a Ptef1-Cas9-TtprC cassette (Additional file 8) and inserted into a p0380-bar plasmid [56] carrying the bar marker to generate the Cas9-expression vector p0380-bar-Ptef1-Cas9-TtprC. A Gibson assembly kit was also used to construct a Cas9-eGFP-expression vector (p0380-bar-Ptef1-Cas9-eGFP-TtprC) in which enhanced GFP (eGFP) was fused to Cas9 as a reporter gene (Additional file 8). To generate sgRNA expression plasmids, an sgRNA scaffold was synthesized by Life Technologies. To express sgRNA in protoplasts, the M. thermophila U6 promoter was amplified from ATCC 42464 genomic DNA using the primer pair U6-F/R (Additional file 7: Table S1) and then fused to the sgRNA scaffold fragment by fusion PCR using U6-F and gRNA-R (Additional file 7: Table S1). The resulting fusion fragment was cloned into a pJET1.2/blunt cloning vector to create the corresponding plasmid U6p-sgRNA (Additional file 9).

To select for specific sgRNAs targeting amdS (GenBank number: M16371.1), cre-1 (MYCTH_2310085), res-1 (MYCTH_2302052), gh1-1 (MYCTH_115968), and alp-1 (MYCTH_2303011), all sgRNA target sites in the genome of M. thermophila were identified using the sgRNACas9 tool [57]. sgRNA target sites with high scores were chosen and the corresponding oligos were ordered (Additional file 7: Table S1). All protospacer sequences used to target the five different genes are presented in Table 1. A target-directed M. thermophila U6 promoter-driven sgRNA was created by overlapping PCR with the primers given in Additional file 7: Table S1 and cloned into a pJET1.2/blunt cloning vector, which yielded the corresponding plasmids U6p-amdS-sgRNA, U6p-cre1-sgRNA, U6p-res1-sgRNA, U6p-gh1-1-sgRNA, and U6p-alp1-sgRNA (Additional file 9).

Donor DNA construction

For construction of gene deletion substrates, the PtrpC-neo cassette was amplified from a p0380-neo plasmid [12]. The 5′ and 3′ flanking fragments of cre-1, res-1, gh1-1, and alp1 were separately amplified from M. thermophila genomic DNA via PCR with paired primers (Additional file 7: Table S1). The amplified 5′, 3′ and PtrpC-neo fragments were assembled and ligated into a pJET1.2/blunt cloning vector using a NEB Gibson assembly kit to generate four donor DNA sequences: donor-cre1, donor-res1, donor-gh1-1, and donor-alp1 (Additional file 10).

Expression of Cas9 in Myceliophthora thermophila

Two Cas9 expression vectors, p0380-bar-Ptef1-Cas9-TtprC and p0380-bar-Ptef1-Cas9-eGFP-TtprC, were transformed into WT M. thermophila strain ATCC 42464 via Agrobacterium-mediated transformation [12]. Colonies grown for 4 days on MM at 35 °C were screened for bar gene resistance using 100 µg mL⁻¹ phosphinothricin (Sigma-Aldrich, St. Louis, MO, USA), followed by sequential identification via PCR analysis. The positive transformants from each construct were named Cas9OE and Cas9-gfp. Cas9OE-positive transformants selected as replicates were subjected to three consecutive rounds of sub-culture; their phenotypes, including secreted protein production, mycelial dry weights and lignocellulosic enzyme activities, were examined in parallel with the WT strain.

Subcellular localization of Cas9-gfp in Myceliophthora thermophila

To localize GFP fusion proteins by microscopy, Cas9-gfp-positive transformants were inoculated in liquid MM containing 2% sucrose as the carbon source and grown for 24 h at 45 °C. Before imaging, hyphae were harvested and incubated with 1 mg mL⁻¹ 4′, 6-diamidino-2-phenylindole for 15 min. The microscopic observation was performed using an Olympus BX51 fluorescence microscopy system, with the ImageJ software used for image processing.

Transformation of Myceliophthora thermophila and M. heterothallica protoplasts

Transformation of M. thermophila and M. heterothallica protoplasts was performed according to a previously described procedure [44]. For amdS mutagenesis, the promoter Ptef1 (MYCTH_2298136) and the full-length acetamidase-encoding gene amdS (GenBank number: M16371.1) were amplified from genomic DNA of M. thermophila and the plasmid p3SR2 using the paired primers Ptef1-F2/R2 and amdS-F/R, respectively. The amdS gene was fused with Ptef1 by overlapping PCR using the primer pair Ptef1-F2/amdS-R. The Ptef1-amdS cassette was then ligated into the EcoRI and HindIII sites of a pCAMBIA-0380 plasmid [56], thereby generating the expression plasmid p0380-Ptef1-amdS. This constructed plasmid was transformed into the WT strain via Agrobacterium-mediated transformation. Colonies grown for 5 days at 35 °C were selected in medium containing 10 mM acetamide as the sole nitrogen source. Positive transformants were identified by PCR with paired primers (Additional file 7: Table S1). From these transformants, the amdS expression strain M1 was chosen and used for further CRISPR/Cas9 manipulation. Briefly, 10 µg of the Cas9-expression PCR cassette bar-Ptef1-Cas9-TtprC and the gRNA expression PCR product U6p-amdS-sgRNA at a molar concentration ratio of 1:1 was co-transformed into protoplasts of the recipient strain M1. After transformation, amdS mutants were inoculated onto MM agar plates supplemented with 2 mg mL⁻¹ FAA and 100 µg mL⁻¹ phosphinothricin. After 3 days incubation at 35 °C, FAA-resistant mutants were isolated and tested for growth on acetamide medium, followed by identification and sequencing via PCR with paired primers (Additional file 7: Table S1).

For single-gene editing, 10 µg of the Cas9-expression PCR cassette bar-Ptef1-Cas9-TtprC, gRNA expression PCR cassette U6p-cre1-sgRNA, and donor-cre1 was mixed at a molar concentration ratio of 1:1:1 and added to the fungal protoplasts. Control experiments were performed by adding 10 µg of donor-cre1 alone, or only the Cas9 cassette and donor-cre1, or only U6p-cre1-sgRNA and donor-cre1 to the fungal protoplasts. Transformants were screened for bar resistance with phosphinothricin (100 µg mL⁻¹) and neo resistance with G418 (40 µg mL⁻¹), followed by PCR identification with paired primers (Additional file 7: Table S1).

For multiple genomic edits, the generated amdS mutant M2 containing a Cas9 expression chassis was used as a host. Multiple genomic modification involving the cre-1, res-1, gh1-1, and alp-1 loci was performed in M2 protoplasts through co-transformation of two, three, or four sets of sgRNA expression cassettes and donor DNA fragments at the same molar concentration. The putative transformants were selected on MM supplemented with 100 µg mL⁻¹ phosphinothricin and 40 µg mL⁻¹ G418, followed by sequential identification via PCR with paired primers (Additional file 7: Table S1).

Determining copy numbers by RT-qPCR

To determine the copy numbers of the integrated PtrpC-neo marker gene in the transformants, the fungal genomic DNAs were extracted as described previously [58] and used as templates for RT-qPCR. The RT-qPCR method was essentially the same as that described by Abad et al. [59]. The qPCR was performed with SYBR Green Real-time PCR Master Mix (TOYOBO, Osaka, Japan) in a CFX96 Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). The primers used for the genes are listed in Additional file 7: Table S1. The actin gene (MYCTH_2314852) was used as an internal control. Each 20 μL reaction contained 1 μL of diluted DNA as the template, 10 μL of 2× SYBR Green Real-time PCR Master Mix, 1 μL of each primer (0.4 M), and 7 μL of H₂O. An equal volume of water instead of DNA was used as a negative control. The qPCR was performed as follows: 95 °C for 30 s, 40 cycles of 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 30 s. A melt curve analysis was performed at the end of each run from 55 to 95 °C with a ramp speed of 0.5 °C to ensure specific sequence amplification of all primers and only one melting temperature on the melting curve. To determine the amplification efficiencies between 90 and 110% of all reactions, the genomic DNA and plasmid DNA samples were diluted serially to construct standard curves and then subjected three times to RT-qPCR.

QL and CT designed the project; QL, RG, and JL carried out the experiments; QL, RG, JL, LL, JZ, and WS analyzed the data; QL and CT wrote the manuscript. All authors read and approved the final manuscript.