Practical guidance for the implementation of the CRISPR genome editing tool in filamentous fungi
- 897 Downloads
Within the last years, numerous reports described successful application of the CRISPR nucleases Cas9 and Cpf1 for genome editing in filamentous fungi. However, still a lot of efforts are invested to develop and improve protocols for the fungus and genes of interest with respect to applicability, scalability and targeting efficiencies. These efforts are often hampered by the fact that—although many different protocols are available—none have systematically analysed and compared different CRISPR nucleases and different application procedures thereof for the efficiency of single- and multiplex-targeting approaches in the same fungus.
We present here data for successful genome editing in the cell factory Thermothelomyces thermophilus, formerly known as Myceliophthora thermophila, using the three different nucleases SpCas9, FnCpf1, AsCpf1 guided to four different gene targets of our interest. These included a polyketide synthase (pks4.2), an alkaline protease (alp1), a SNARE protein (snc1) and a potential transcription factor (ptf1). For all four genes, guide RNAs were developed which enabled successful single-targeting and multiplex-targeting. CRISPR nucleases were either delivered to T. thermophilus on plasmids or preassembled with in vitro transcribed gRNA to form ribonucleoproteins (RNPs). We also evaluated the efficiency of single oligonucleotides for site-directed mutagenesis. Finally, we were able to scale down the transformation protocol to microtiter plate format which generated high numbers of positive transformants and will thus pave the way for future high-throughput investigations.
We provide here the first comprehensive analysis and evaluation of different CRISPR approaches for a filamentous fungus. All approaches followed enabled successful genome editing in T. thermophilus; however, with different success rates. In addition, we show that the success rate depends on the respective nuclease and on the targeted gene locus. We finally present a practical guidance for experimental considerations aiming to guide the reader for successful implementation of CRISPR technology for other fungi.
KeywordsFilamentous fungi Cell factory Thermothelomyces thermophilus Myceliophthora thermophila CRISPR Genome editing Cas9 Cpf1 Cas12a RNP Multiplexing Selection-free gene targeting
CRISPR from Prevotella and Francisella
Clustered Regularly Interspaced Short Palindromic Repeats
Six million fungal species are estimated to exist on Earth , but we know only about 100,000 . Most are saprophytes; however, many pose a threat for other organisms including man. Only a few are exploited in biotechnology as cell factories. Aspergillus niger has been the pioneer fungus of modern biotechnology and used for exactly 100 years for the production of citric acid and since then together with other fungal cell factories for many other products including organic acids, enzymes, drugs, antibiotics and vitamins to name but a few [3, 4]. To improve our understanding on fungal biology underlying pathogenicity or metabolic capabilities, fast and efficient genetic manipulation tools are a fundamental prerequisite.
Plasmid-based and RNP-based
Plasmid-based and RNP-based
Plasmid- based and RNP-based
To overcome this issue, we tested in the current study the efficiency of three different nucleases for gene targeting in the cell factory T. thermophilus. This filamentous fungus is of current research interest because it exhibits a large capacity for plant biomass degradation and represents a potential reservoir of novel enzymes for many industrial applications. It was formerly known as Myceliophthora thermophila and a CRISPR method based on Cas9 has been published for this cell factory in 2017 . The market for enzymes is huge with a total value of approximately $ 4 billion in 2018 . The market leader was Novozymes, with a share of 48%, followed by Danisco (21%), DSM (6%), AB Enzymes (5%) and BASF (4%). Within this market, household care enzymes made up 32% of sales, closely followed by food and beverage enzymes (29%), bioenergy (19%), agricultural and feed (14%) and other technical and pharma enzymes (6%). The T. thermophilus strain ATCC 42464 is predominantly used in academic research groups as the general wild-type strain. For industrial use, the proprietary mature enzyme production strain C1 was developed . The main features of strain C1 are production levels up to 100 g/L protein, and the maintenance of low viscosity levels during fermentation.
We present here a comprehensive survey of different CRISPR gene-targeting approaches for the T. thermophilus strain ATCC 42464 including the successful implementation of two new Cpf1 nucleases. We tested the Cpf1 nucleases from Francisella novicida (FnCpf1) and Acidaminococcus sp. (AsCpf1) to broaden the genome editing toolbox and compared their performance to the well-established Cas9 nuclease from Streptococcus pyogenes (SpCas9). Note that the recognition sequence for FnCpf1 is 5′‐TTN‐3′ and 5′‐TTTN‐3′ for AsCpf1, whereas SpCas9 recognizes 5′‐NGG‐3′ . Previous studies have shown that the genome editing efficiency can be different between AsCpf1 and FnCpf1. AsCpf1 performed better in human cell lines , whereas genome editing with FnCpf1 was more efficient in S. cerevisiae . Single, double, triple and quadruple gene-targeting were successfully established in T. thermophilus and the efficiency of a plasmid-based or RNP-based provision of the respective nucleases compared. We finally optimized transformation protocols for both approaches with respect to efficiency and scalability.
Results and discussion
RNP application of FnCpf1, AsCpf1 and SpCas9 for single targeting
Strains of filamentous fungi that are deficient in the non-homologous end joining (NHEJ) pathway, i.e. with reduced ectopic integration events during transformation, are preferred as hosts for efficient genome editing due to their high frequencies of DNA integration via homologous recombination . In the case of T. thermophilus, this was recently proven for the ku70 gene, which is a central element of the NHEJ machinery. Its inactivation resulted in a threefold higher homologous recombination rate . Therefore, we have deleted another central element of the NHEJ machinery, the predicted ku80 ortholog, (MYTH_2118116), in the wild-type T. thermophilus strain ATCC42464 using SpCas9 and amdS as selection marker (for details see “Methods”). Correct deletion of ku80 was verified in strain MJK19.4 by diagnostic PCR and Southern blot analysis (Additional file 1 and data not shown). This strain was selected for removal of the amdS gene via FAA counterselection (see “Methods”) resulting in strain MJK20.2.
Transformants and gene deletion efficiency targeting the pks4.2 gene
No. of colonies analysed
Editing efficiency (%)
RNP application of FnCpf1, AsCpf1 and SpCas9 for multiplex-targeting
To investigate whether the three CRISPR nucleases support targeting of up to four genes simultaneously, three of which via a selection-free process, we performed the following strategy: (i) deletion of the pks4.2 gene (using amdS as selection marker) resulting in an easy detectable colour mutant (Fig. 1a), (ii) replacement of the endogenous snc1 gene (MYTH_64173) with a functional snc1::eGFP fusion construct for detection of GFP fluorescence via confocal microscopy (note that snc1 encodes a SNARE protein and is an established marker for secretory vesicles in filamentous fungi (Fig. 1b, ), (iii) deletion of the alp1 gene (MYTH_2303011) encoding an alkaline protease which was previously shown to become successfully targeted by SpCas9 in T. thermophilus , and (iv) deletion of a non-verified protein encoding a predicted transcription factor. For brevity, we named it ptf1 in this study. Donor DNAs were provided for all genes and details can be found in “Methods” section.
Transformants and PCR-confirmed editing efficiency for the simultaneously targeted alp1, pks4.2, snc1, and ptf1 gene loci
No. of colonies
Editing efficiency (%)
Comparison of RNP-based and plasmid-based application of FnCpf1 for multiplex-targeting
Editing efficiency for four different gene loci of FnCpf1
No. of analysed transformants
No. of analysed transformants
SON-based targeting of FnCpf1 and SpCas9
Single-stranded oligonucleotides (SONs) have been shown to be efficient templates for the repair of SpCas9 and LbCpf1 (from Lachnospiraceae bacterium) induced DNA double-strand breaks in NHEJ-deficient A. nidulans and A. niger [9, 24]. We therefore tested whether this approach (which can be harnessed to introduce specific point mutations into the locus of interest) can also be followed using FnCpf1 and SpCas9 nucleases in T. thermophilus. We thus applied 90 bp long oligonucleotides homologous to part of the pks4.2 locus which were designed to introduce three stop codons in the centre part (Additional file 3). The selection marker was present on the donor DNA for the second target, pks4.1 (plasmid pMJK22.19). In total, 30 (25) amdS expressing transformants were identified for FnCpf1 (SpCas9), 5 (3) of which displayed the respective spore colour change indicative for a pks4.2 gene inactivation. All eight transformants were picked and sub-cultivated. Correct integration of the pks4.1 donor DNA was verified by PCR and the respective pks4.2 locus PCR amplified and sequenced. The sequencing results verified that all 8 transformants were successfully targeted by both nucleases and that the SON introduced the desired gene edits (Additional file 3). For the first time, this data provides evidence that an oligonucleotide-mediated repair approach can be followed in T. thermophilus for site-directed mutagenesis applying either FnCpf1 or SpCas9.
MTP-based method for high-throughput gene targeting
Practical guidance for the implementation of CRISPR technology in filamentous fungi based on data obtained for T. thermophilus in this study
Preparation of nuclease
Cloning of the nuclease into a plasmid prior transformation is mandatory. When constitutively expressed, risk of off-targets should be considered. When present on AMA-plasmid, the risk should be lower but still present
Cloning of the nuclease into a plasmid allowing heterologous expression, e.g. in E. coli, is a prerequisite. Once established and purified, the nuclease can be aliquoted and stored prior to use. As the protein does not become expressed in the targeted fungus, the risk of off-targets should be very small
Preparation of guide RNA
Plasmid-based, thus more stable during handling and storage
Involves in vitro transcription, hence potentially sensitive to handling errors
Easy but requires preassembly of RNPs
Very high also with four targets
Very high for single and double targets
Low for three and four targets
Single-targeting efficiency of FnCpf1, AsCpf1, SpCas9
Multiplex-targeting efficiency of FnCpf1
High (34 % ± 6 % in this study)
Low (13 % ± 2 % in this study)
MTP-based down-scaling for FnCpf1
Possible with no loss in efficiency with respect to single and double targeting
Possible with no loss in efficiency with respect to single targeting*
Microbial strains and cultivation conditions
Fungal strains used in this study are given in Additional file 4. Strain MJK20.2 was used as progenitor isolate as this strain is deficient in the non-homologous end joining pathway (Δku80), reducing ectopic integration events during transformation and thus enabling targeted integration . Strains were grown at 37 °C in minimal medium (MM) or complete medium (CM), consisting of MM supplemented with 1% yeast extract and 0.5% casamino acids . All bacterial plasmids were propagated in Escherichia coli DH5α using 100 µg/mL ampicillin or 50 µg/mL kanamycin for selection.
All molecular techniques were performed according to standard procedures described previously . T. thermophilus transformation and genomic DNA extraction were performed as described elsewhere . When required, plates were supplemented with acetamide (15 mM) and caesium chloride (10 mM). Primers and plasmids used in this study are given in Additional files 5 and 6, respectively. All plasmids were sequenced and will be made available on reasonable request. Strain MJK20.2 (Δku80) was generated as follows: ku80 was deleted in the wild type strain ATCC42464 with FnCpf1 or SpCas9 using PCR-amplified split marker fragments containing the amdS marker and about 1.2 kb flanks each for homologous integration. The 3′ split marker fragment contained the 5′ flank to mediate a fast removal of the amdS marker. The resulting strain MJK19.1 was sub-cultivated on FAA medium plates to obtain the marker-free Δku80 strain MJK20.2. Strains were analysed by Southern blot analysis to verify correct integration of the fragments and removal of the marker gene (Additional file 1). For all other targets, donor DNA with flanks of about 1 kb length each were used. The amounts of donor DNA are specified in the RNP-based and plasmid-based approaches described below.
Genome editing using RNP-based approach
The plasmid containing the expression cassette for SpCas9 (pET28a/Cas9-Cys) was obtained from addgene (#53261). T. thermophilus codon optimized FnCpf1 and AsCpf1 was cloned into plasmid pET28a giving plasmid pMJK16.1 and pMJK17.1, respectively. E. coli strain Rosetta™ 2(DE3)pLysS (Novagen) was freshly transformed with the respective expression plasmids. Four mL of TB medium (12 g/L tryptone, 24 g/L yeast extract, 5 g/L glycerol, 2.31 g/L KH2PO4, 12.54 g/L K2HPO4) plus 50 µg/mL kanamycin and 20 µg/mL chloramphenicol were inoculated from a single colony and incubated at 37 °C and 250 rpm overnight. 400 µL of these precultures were used to inoculate 40 mL of TB medium including antibiotics, which was incubated at 37 °C and 250 rpm until an optical density (OD600) of 5.0–8.0 was reached (approximately 5–7 h). Main cultures (1 L in 5 L Erlenmeyer flasks) with TB medium, autoinduction solution (5 g/L glycerol, 0.5 g/L glucose, 2 g/L α-lactose monohydrate) and the corresponding antibiotic(s) were inoculated with these 40 mL cultures to an OD600 of 0.1 and incubated in shake flasks at 37 °C and 160 rpm for 2 h. Afterwards, the temperature was decreased to 18 °C and the cells cultivated for at least 18 up to a maximum of 40 h. Proteins were purified as described previously  using Ni–NTA resin (Qiagen Germany).
Target sequences were selected in silico using Cas-Designer and Cas-OFFinder (http://www.rgenome.net/cas-offinder/) . Respective gRNAs including PAM sites were generated as described before . In brief, gRNAs were in vitro transcribed using the T7 promoter with an additional ATG at the front (ATGTAATACGACTCACTATAGG). For sequence information, see Additional file 5.
RNP assembly was done as described previously  with the following modifications. Prior to the transformation into fungal protoplasts, RNP complexes were assembled containing 30 µg CRISPR nuclease (5 µL), 2 µL 10 × Cas9 activity buffer, 1 µL gRNA and 12 µL of nuclease-free water in a 1.5 mL reaction tube. The mixture was incubated at 37 °C for 15 min to allow RNP complex formation. For multiplex-targeting, each target RNP complex was formed separately. For each transformation, 100 µL protoplasts, 10 µL donor DNA (5 µg), 20 µL RNP complex (up to 80 µL for multiplex-targeting), 20 µL 2× STC, 25 µL 60% PEG 4000 buffer and 20 µL 10 × Cas9 activity buffer were mixed in a 50 mL Greiner tube. Transformations with plasmid MT28 and/or sterile water served as controls. Note that this protocol differed from  with respect to PEG: 60% PEG 4000 was used in this study instead of 25% PEG 6000. Transformants were sub-cultivated twice on medium with 15 mM acetamide as nitrogen source. Genomic DNA was extracted from transformants. Insertion of the donor cassette at the respective locus was confirmed by diagnostic PCR.
Genome editing using plasmid-based approach
3 µg of the fncpf1 encoding plasmid MT2286 was co-transformed with 2 µg of each plasmid DNA encoding respective gRNAs separated by direct repeats (e.g. pMJK31.1 for pks4.2 & snc1 gRNA) as described by  and 3 μg donor DNA into T. thermophilus as follows: 100 µL protoplasts (~ 5 × 106 protoplasts), 10 µL total DNA and 25 µL 60% PEG 4000 buffer were mixed in a 50 mL Greiner tube at room temperature. For transcription of the gRNA the U6 promoter was used. Expression of fncpf1 was done according to [34, 12]. For codon optimisation the most frequent codons were used . Transformations with plasmid MT28 and/or sterile water served as controls. Note that this protocol differed from  with respect to PEG: 60% PEG 4000 was used in this study instead of 25% PEG 6000. Transformants were sub-cultivated twice on medium with 15 mM acetamide as nitrogen source. Genomic DNA was extracted from putative transformants. Insertion of the donor cassette at the respective locus was confirmed by diagnostic PCR.
Genome editing using SON-based approach
SON-based donor DNA targeting pks4.2 was designed with 35/32 bp (up-/downstream) homologous arms containing 3 stop codons. For sequence information, see Additional file 3. Selection was based on the Δpks4.1 deletion cassette (pMJK22.19), hence a double targeting approach was followed: RNP complexes were assembled containing 30 µg FnCpf1 (5 µL), 2 µL 10 × Cas9 activity buffer, 1 µl gRNA and 12 µL of nuclease-free water in a 1.5 mL reaction tube. The mixture was incubated at 37 °C for 15 min to allow RNP complex formation. For each transformation, 100 µL protoplasts, 5 µL donor DNA (5 µg), 10 μL SON (100 μM stock solution), 40 µL RNP complex, 20 µL 2× STC, 25 µL 60% PEG 4000 buffer and 20 µL 10 × Cas9 activity buffer were mixed in a 50 mL Greiner tube. Note that this protocol differed from  with respect to PEG: 60% PEG 4000 was used in this study instead of 25% PEG 6000. Transformants were sub-cultivated twice on medium with 15 mM acetamide as nitrogen source. Genomic DNA was extracted from transformants. Insertion of the donor cassette at the respective locus was confirmed by diagnostic PCR.
Genome editing using MTP-based approach
The volume for the transformation reaction was reduced to 200 μL and transformation was done in a 1.5 mL reaction tube. Both, freshly prepared and cryopreserved protoplasts have been used. For one transformation using the plasmid approach, 10 μL protoplasts (~ 5 x 105) were mixed with 1 μL donor DNA (1 μg), 1 µL FnCpf1 (1 μg), 1 μL gRNA plasmid (1 μg) and 2.5 μL 60% PEG 4000 buffer at room temperature. Afterwards 61.5 μL 60% PEG 4000 buffer was added and exactly five minutes later 123 µL STC was added. Instead of using top agar to distribute cells on an agar plate (15 cm diameter), the 200 μL protoplast mix was spread onto a small plate (9 cm diameter). For the RNP approach, 5 μL of RNP mixture was added from a 20 μL RNP complex reaction mixture. Identification and analysis of transformants were performed as described above.
Cryopreservation of protoplasts
350–500 μL protoplasts (~ 1 × 107) were mixed 1:1 with 20% Polyvinylpyrrolidone 40 solved in STC buffer. This mixture was deep frozen at − 80 °C using isopropanol for − 1 °C/min freezing. Before transformation, frozen protoplasts were washed with 10 mL cold STC buffer and spun down for 5 min at 1500 rpm and 4 °C. Protoplasts were resuspended with cold STC and used for transformation.
In vitro SpCas9 cleavage assay
In brief, the plasmids with either the original Psnc1::gfp::snc1 sequence or the SpCas9 PAM site mutated sequence were used as donor DNAs. Each 600 ng were restricted with 10 U NotI in a total volume of 20 µL to generated linearized DNAs. After heat-inactivation (20 min at 80 °C) of NotI, the mixtures were immediately added without further purification to a 30 µL reaction mixture containing 1 µL gRNA and 1 µL Cas9 protein. After 60 min incubation at 37 °C, the reaction was quenched by adding 3 µL 0.5 M EDTA and 7 µL 6× gel loading dye. Samples were incubated for 15 min at 65 °C and analysed by electrophoresis on a 1% agarose gel.
Genotypic, phenotypic and microscopic screens of CRISPR transformants on agar media
Putative transformants were analysed as follows: Δpks4.2 transformants were sub-cultivated three times on MM agar medium and the spore colour formation compared to the wild-type strain. In case of questionable phenotypes, strains were further subjected to diagnostic PCR. Integration of Psnc1::gfp::snc1 was analysed using fluorescence microscopy. In brief, colonies cultivated on selective agar MM medium containing 15 mM acetamide for 24 h at 37 °C and fluorescence images were taken using an inverted TCS SP8 (Leica, Germany) as described earlier . Most colonies with GFP-secretory vesicle signals had correct integration of the donor DNA at the snc1 locus as checked by diagnostic PCR (~ 99%). Consequently, transformants with GFP- secretory vesicle signals, were considered snc1 targeted. For Δpks4.1, Δalp1 and Δptf1 diagnostic PCR was done on the corresponding locus. For primer sequence information, see Additional file 5.
We acknowledge support by the German Research Foundation and the Open Access Publication Funds of TU Berlin. We thank BASF SE for financial support.
MJK designed gene editing primers, generated und purified CRISPR enzyme variants, generated T. thermophilus mutants, and conducted PCR, Southern, phenotypic and microscopic analyses. SS and SH provided the constructs for genome editing using the plasmid-based approach. VM initiated this study, coordinated the project and co-wrote the final text. MJK, TS, SS and SH were involved in discussions and contributed to writing the manuscript. All authors read and approved the final manuscript.
Ethics approval and consent to participate
Consent for publication
The authors declare they have no competing interests.
- 2.Helen A. Ainsworth and Bisby’s dictionary of the fungi (10th edition). Ref Rev. 2009;23:42.Google Scholar
- 13.Winter AB, Steel AL. The Novozymes Report 2018. https://report2018.novozymes.com.
- 14.Visser H, Joosten V, Punt PJ, Gusakov AV, Olson PT, Joosten R, et al. Development of a mature fungal technology and production platform for industrial enzymes based on a Myceliophthora thermophila isolate, previously known as Chrysosporium lucknowense C1. Ind Biotechnol. 2011;7:214–23.CrossRefGoogle Scholar
- 38.Katayama T, Nakamura H, Zhang Y, Pascal A, Fujii W, Maruyama JI. Forced recycling of an AMA1-based genome-editing plasmid allows for efficient multiple gene deletion/integration in the industrial filamentous fungus Aspergillus oryzae. Appl Environ Microbiol. 2019;85:1–16.Google Scholar
Open AccessThis 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.