2.1. CRISPR-Cas9 system

The CRISPR-Cas9 system consists of the Cas9 nuclease and small guide RNA (sgRNA). Numerous cells, animals, and plants have been edited using this potent and promising gene editing system (Huang et al., 2022). The CRISPR-Cas9 system (Figure 3A), which is derived from Streptococcus pyogenes (SpCas9), is one of the most characterized, most commonly utilized, and most active Cas enzymes among the CRISPR-Cas systems (Hsu et al., 2013). SpCas9 is a large protein (1,368 amino acids) with two distinct lobes in its apo state: the alpha-helical recognition (REC) lobe, composed of three alpha-helical domains (Hel-I, Hel-II, and Hel-III), while the nuclease (NUC) lobe retains the conserved HNH, the split RuvC nuclease domains, and the more variable C-terminal domain (CTD). The elongated CTD also displays a Cas9-specific fold and contains PAM-interacting sites required for protospacer-adjacent motifs (PAM) interrogation (Jiang et al., 2017). The sgRNA is designed by fusing crRNA with trans-activating crRNA (tracrRNA). It is an RNA complex containing a human-designed 20-bp nucleotide sequence complementary to the target gene at the 5′ end, followed by a repeat of crRNA and an inverted repeat of tracrRNA (Prasad et al., 2021). During gene editing, a ribonucleoprotein composed of the Cas9 protein and sgRNA recognizes the PAM and target sequence on the gene. The adjacent DNA sequence unwinds and forms a DNA-sgRNA complex (R loop). The HNH and RuvC domains of the Cas9 protein cut the complementary and non-complementary strands of the sgRNA, respectively, to generate double-strand breaks (DSBs). These breaks are then repaired through non-homologous end joining (NHEJ) or homology-directed repair (HDR) (Szczelkun et al., 2014).

Open in a separate window

Figure 3

Gene editing schematic of four commonly used CRISPR-Cas systems (Cas9, Cas12a Cas13a and AsCas12f). (A) The CRISPR-Cas9 system is capable of cleaving double-stranded DNA (dsDNA) 3bp upstream of the protospacer adjacent motif (PAM) under the guidance of a single guide RNA (sgRNA), resulting in a blunt double-stranded breaks (DSBs) at the target site. (B) The CRISPR-Cas12a system relies on crRNA to recognize the PAM of dsDNA, which then leads to the cleavage of the target DNA and the production of sticky-ended DSBs. (C) The CRISPR-Cas13a system utilizes crRNA to specifically target and cleave single-stranded RNA downstream of the PFS. (D) The CRISPR/AsCas12f system has the unique ability to cleave single-stranded DNA (ssDNA) without the need for PAM. Additionally, it is capable of cleaving dsDNA to create sticky-ended DSBs, which is achieved through the recognition of PAM sites. (E) Base editors can generate C·G to T·A, A·T to G·C, C·G to G·C point mutations without creating DSB. (F) Schematic diagram illustrating the components involved in the PE mechanism.

PAM is essential for the Cas9 protein to recognize and cleave target genes. However, the specific PAM sites recognized by Cas9 proteins vary among different sources. For instance, SaCas9 from Staphylococcus aureus recognizes NNGRRT (where N can be any nucleotide and R can be either A or G) (Ran et al., 2015), NmeCas9 from Neisseria meningitidis recognizes NNNNGATT (Zhang et al., 2015), and SpCas9 from Streptococcus pyogenes recognizes the simpler PAM sequence NGG (Anders et al., 2014). However, the requirement for Cas9 to recognize specific PAM sites during gene editing not only limits the locations where editing can be performed but also affects the efficiency and flexibility of the editing process. To address these limitations, scientists have developed various Cas9 variant systems that can extend PAM compatibility, such as xCas9, SpCas9-NG, SpG, and more (Hu et al., 2018; Nishimasu et al., 2018; Miller et al., 2020; Walton et al., 2020). One of the most recent advances is a “PAMless” Cas9 mutant protein called SpRY, which significantly increases the editable range of Cas9 and can identify practically all PAM sequences (Walton et al., 2020).

2.2. CRISPR-Cas12 system

Cas12 and Cas9 are the two commonly used CRISPR-Cas systems. However, Cas12 is generally smaller in size compared to Cas9 (Figure 4). During gene editing, CRISPR-Cas12 can recognize double-stranded DNA (dsDNA) rich in thymine (T) PAM sites only under the leadership of crRNA, and then accurately cut the desired sequence. Additionally, Cas12 also has target-specific trans-cleavage activity, allowing it to non-specifically cleave single-strand DNA (ssDNA) (Li et al., 2018) (Figure 3B). Following the identification of Cas12a (Cpf1) (Zetsche et al., 2015), scientists have successively developed several miniaturized Cas proteins such as Cas12b (C2c1) (Ming et al., 2020), Cas12e (CasX) (Liu J. J. et al., 2019), Cas12i (Yan et al., 2019), Cas12j (CasΦ) (Pausch et al., 2020), Cas12l (Sun et al., 2023), Cas12f (Cas14) (Harrington et al., 2018; Karvelis et al., 2020; Wu et al., 2021; Kim et al., 2022), among which the CRISPR-Cas12 genome editing technology has been advanced rapidly.

Open in a separate window

Figure 4

Cas9, Cas12 and Cas13 protein types and sizes.

It is worth noting that Cas12f, originally classified as Cas14, was previously thought to only function as an endonuclease capable of specifically cutting desire ssDNA and other ssDNA non-specifically independent of PAM sequences (Harrington et al., 2018) (Figures 3D1,D2). Nevertheless, recent research has shown that Cas12f also has PAM-dependent targeting activity towards dsDNA, which differs from previous studies (Karvelis et al., 2020). Compared to the Cas9 protein, the wild-type Cas12 protein has lower gene editing efficiency and is more likely to produce deletion products (Xin et al., 2022). Recently, an engineered AsCas12f system known as “enAsCas12f” has been designed to increase the editing efficiency of the Cas12 protein. This system has a cleavage activity of up to 11.3 times that of the wild-type AsCas12f, although it is only a third as big as SpCas9. Due to this, it is among the most effective and small-sized CRISPR systems available, introducing a new area of CRISPR-based gene editing (Wu et al., 2023).

3.1. Expression of the CRISPR-Cas system

There are three main methods to express the CRISPR-Cas system in filamentous fungi, and the expression strategies of Cas9 and Cas12 are identical as well. The first method involves transforming a plasmid containing both the sgRNA transcription sequence and the Cas9 protein expression sequence (Wang et al., 2020). This allows for continuous expression of both the Cas protein and sgRNA in the fungi, increasing the likelihood of successful cleavage or targeting of the target DNA/RNA. However, this approach is dependent on utilizing promoters to express the Cas protein and sgRNA in vivo. The second method involves transferring the sgRNA to be transcribed in vitro into a strain expressing the Cas protein (Liu et al., 2015). This method has the benefit of not requiring in vivo promoters for sgRNA expression, but it comes with extra work and the danger of sgRNA degradation. A third strategy is to transform ribonucleoproteins (RNP) assembled from Cas and sgRNA into the cells (Jan et al., 2019; Li et al., 2019). This method is not limited by Cas9 translation or sgRNA transcription levels, unlike the previous methods, and can be used in filamentous fungi that lack a codon usage database and cannot be codon-optimized for the host. However, directly converting RNP into fungal protoplasts presents the problem of low efficiency, which to some extent limits its application in fungi. According to recently published studies, extending the period of incubation and adding the chemical reagent Triton X-100 can significantly improve RNP’s conversion performance (Wu et al., 2020).

3.2. Delivery of the CRISPR-Cas9 system and screening markers

Agrobacterium-mediated transformation (AMT) and polyethylene glycol-mediated transformation (PMT) are the two primary integration methods for the CRISPR-Cas system vectors, which are frequently employed for genome editing in filamentous fungi (AMT) (Song et al., 2019). By employing PEG induction to introduce foreign DNA or RNA fragments into protoplasts, PMT causes these fragments to randomly integrate within the fungal genome, creating multiple copies (Yu et al., 2014). On the flip side, AMT utilizes the expression of Virulence genes (Vir) on the Agrobacterium tumor-inducing (Ti) plasmid to transfer DNA (T-DNA) and exogenous DNA into the mycelium, spores, or protoplasts of filamentous fungi, resulting in single copy integration (Sayari et al., 2019). Additionally, electroporation has been successfully used to transform plasmid vectors carrying the Cas9 gene into Alternaria alternata (Min et al., 2016; Vyas et al., 2018).

The screening of transformants requires the use of available markers. However, the selection of markers for filamentous fungal screening is limited. Common types of markers include resistance marker genes, nutrient deficiency genes, and phenotype reporter genes. Resistance genes, such as the hygromycin B resistance gene (hphR/hygR), glufosinate resistance gene (bar), and geneticin resistance gene (G418), have been effectively employed to screen positive strains. For example, filamentous fungi selected for hph/hyg as a positive screening marker include Aspergillus fumigatus (Fuller et al., 2015), Aspergillus niger (Kuivanen et al., 2019), Talaromyces atroroseus (Nielsen et al., 2017), Pochonia chlamydosporia (Youssar et al., 2019), Leptosphaeria biglobosa (Darma et al., 2019), Fusarium fujikuroi (Shi et al., 2019), Fusarium oxysporum (Wang et al., 2018), etc. The bar selection includes Cordyceps militaris (Chen et al., 2018), Thermobrachicum celere (Liu et al., 2017), Neurospora crassa (Matsu-Ura et al., 2015), and Pyricularia oryzae (Arazoe et al., 2015). G418 antibiotics are used to screen for positive filamentous fungi, such as Ashbya gossypii (Jiménez et al., 2019), Ustilaginoidea virens (Liang et al., 2018), and Phytophthora sojae (Miao et al., 2018). In addition to resistance genes, the expression of trophic markers, such as the acetamidase encoding gene (amds) and orotidine-5′-phosphate decarboxylase gene (ura3), is also an effective means to screen for positive filamentous fungi strains. For example, scientists have expressed amds genes derived from Aspergillus nidulans in Aspergillus niger (Zheng et al., 2019) and Penicillium chrysogenum (Pohl et al., 2016), giving these filamentous fungi the capacity to survive on medium with acetamide as the only nitrogen source, while wild-type strains cannot grow. Pyr4 gene deficient strains selected by 5-FOA were also used as receptor strains for gene expression, and the researchers established screening markers for uridine deficiency and applied them to filamentous fungi such as Mucor circinelloides (Nagy et al., 2019) and Aspergillus fumigatus (Zhang et al., 2016). Enhanced green fluorescent protein (EGFP) and red fluorescent protein (RFP) are excellent indicators for isolating filamentous fungi by fluorescence microscopy or flow cytometry (Nodvig et al., 2015). Some scientists evaluated Cas expression levels by constructing Cas with eGFP (Liu et al., 2017; Chen et al., 2018).

3.3. NHEJ and HDR-mediated gene knockout

In filamentous fungi, CRISPR-Cas editing of genes is primarily accomplished via the NHEJ and HDR pathways to repair double-strand breaks (DSBs) caused by Cas protein splicing. The NHEJ repair pathway can occur throughout the cell cycle, with a major role played by the ku70 and ku80 proteins in the G1 phase (Sfeir et al., 2015). HDR is typically active in the S or G2 phase of the cell cycle, as sister chromatids provide homologous donor DNA fragments at this stage (Sfeir et al., 2015). Without the presence of a donation template DNA, the DSB ends produced by Cas9 are primarily joined together through NHEJ, which can result in frameshift mutations due to random base insertion or deletion. This ultimately leads to target gene inactivation or mutation. NHEJ-mediated gene knockout is frequently employed in the initial stages of the CRISPR-Cas system’s utilization to disrupt endogenous marker genes and verify the system’s construction success. It can also be employed for gene inactivation to reveal gene function (Bauer et al., 2015), as demonstrated in studies on Aspergillus fumigatus (Fuller et al., 2015), Magnaporthe oryzae (Li et al., 2019), Fusarium fujikuroi (Shi et al., 2019).

However, the deletions or insertions introduced by NHEJ at the cleavage sites were random and did not result in the changes expected. In contrast, HDR is a more effective method because it provides artificial DNA repair templates, which enable the exact introduction of mutations or sequences at the target spot. The performance of HDR-mediated gene editing is influenced by factors such as the distance between the homology arm and the DSB, the size of the homology arm, and the type of donor DNA fragment used. For example, Dong et al. found that in their research on Aspergillus niger, the closer the homologous arm was to the double-strand break, the higher the gene editing efficiency. They designed three donor DNA fragments containing the resistance screening marker gene hygB, with homologous arms of the same size at both ends but at different distances from the DSB (0kb, 1kb, and 5kb). The resulting hygB integration efficiencies were 80, 50, and 10%, respectively (Dong et al., 2019). Liu et al. effectively implemented a CRISPR-Cas9 system in T. reesei by means of the introduction of homologous arms with different lengths around the selectable marker. They demonstrated that the addition of a pair of 200-bp homology arms resulted in a homologous recombination frequency of approximately 93%, allowing for CRISPR-Cas9-mediated gene knockout in T. reesei (Liu et al., 2015). In addition to linear donor DNA, successful gene editing has also been achieved in Aspergillus oryzae (Katayama et al., 2019) and Trichoderma reesei (Liu Q. et al., 2019) using circular donor DNA. Compared to linear DNA donors, circular DNA donors have been found to be more efficient in gene editing, likely due to the reduced integration into filamentous fungi chromosomes without a DSB, resulting in a lower false positive rate during screening (Chen et al., 2017). In recent years, single-strand oligonucleotides have also been used as homologous recombination donors for CRISPR-Cas-mediated genome editing in filamentous fungi, and research has shown that even 60-mer single-strand oligonucleotides can accurately alter target genes in Aspergillus niger employing the CRISPR-Cas system (Kun et al., 2020).

3.4. Multiple editing

Some fungal traits and natural products are usually regulated by multiple genes, making it necessary to simultaneously knock out target genes in practical applications. Yet, the capacity of conventional techniques such as ZFN and TALEN to simultaneously knock out several genes is hampered. Whereas, by expressing several sgRNAs in filamentous fungi, CRISPR-Cas technology enables simpler multigene editing. There are three main strategies for expressing multiple sgRNAs in filamentous fungi. The first method is to introduce a vector carrying the Cas9 expression sequence into the host fungi through the AMT method. Following that, introduce several mature sgRNAs generated in vitro into the Cas9-positive cells through the PMT method (Liu et al., 2015). This method has been successfully used in Aspergillus fumigatus (Zhang et al., 2016), Aspergillus niger (Kuivanen et al., 2016), and Beauveria bassiana (Chen et al., 2017). The second method involves using the tRNA-spacer system to express multiple sgRNAs. The sgRNAs are concatenated and interspaced by 5′ and 3′ splice sites of a tRNA, which are recognized by RNase P and RNase Z. This allows for the release of mature sgRNAs through the action of these enzymes (Phizicky et al., 2010). Employing polymerase III promoters and tRNA spacers, Nodvig et al. (2018) designed a vector system that can deliver Cas9 and several sgRNAs. With this approach, two point mutations and one gene insertion have been introduced in a single transformation experiment with great success and excellent performance. Chen et al. (2022) have developed a marker-free CRISPR-Cas9-TRAMA genome editing system by utilizing an endogenous tRNA-processed element, allowing for several gene-specific editing as well as large synthetic cluster elimination in Cordyceps militaris. It is worth noting that a trancrRNA is not required for the single-RNA-guided nuclease Cas12a to function. The capacity of Cas12a to use specialized CRISPR arrays to encode two or more crRNAs in a single transcript gives it an essential benefit over Cas9. This contains multiple crRNAs separated by short direct repetitions, making it simple to multiplex the editing of genes and regulation (Liu Q. et al., 2019). The third method involves employing a linker sequence to bind two distinct sgRNA cassettes together so that they are expressed under the same promoter and terminator. However, research has revealed that this strategy is less effective at editing than driving distinct sgRNAs by independent promoters (Kearns et al., 2015).

The lack of available genetic markers is a major obstacle in the process of multigene editing in filamentous fungi. However, the development of CRISPR-Cas-assisted marker recycling technology (Camr technology) has reduced the need for multiple screening markers in experiments. An example of this method is the construction of hendecuple mutants M. thermophila, achieved through three successive transformations using two selectable markers, neo and bar (Liu Q. et al., 2019). In the first round, a triple mutant was created using transient CRISPR-Cas12a/Cas9-mediated homology-directed repair (HDR) (Figure 6). The neo cassette replaced an endogenous locus, while two target genes were deleted without the use of markers. In the second round, the neo cassette was removed and the bar was inserted into a new endogenous locus through HDR using the CRISPR-Cas system. Additionally, markerless gene disruption and seamless gene substitution were also applied to two more gene loci. The octuple mutant was selected in the third round to undergo additional genetic modification through the CRISPR technique. The selectable marker neo was inserted into a new endogenous locus, while the marker bar was seamlessly removed without leaving a scar. Another new gene also underwent markerless deletion. This allows for the reuse of the marker bar in subsequent transformation rounds. Moreover, the use of autonomously replicating plasmids with MAM1 sequences allows for the repeated use of selectable marker genes during transformation (Figure 7). This is because genes on such plasmids are not easily integrated into fungal genomes, and the plasmids are easily lost under non-selective culture conditions (Katayama et al., 2019; Wang et al., 2019; Meng et al., 2022).

Open in a separate window

Figure 6

Schematic strategy of CRISPR–Cas-mediated marker recycling approach for iterative multiplex genome editing in M. thermophila. The hendecuple mutants were created through a series of three transformations, utilizing the selectable markers neo and bar. In the first transformation, a triple mutant was generated usingtransient CRISPR-Cas12a/Cas9-mediated homology-directed repair (HDR). This involved replacing an endogenous locus with the neo cassette and deleting two target genes without the use of markers. In the second transformation, the neo insert was removed and bar was used to replace another endogenous locus in the triple-mutant strain through HDR with the CRISPR-Cas12a/Cas9 system. Additionally, two other gene loci underwent seamless gene replacement and markerless gene disruption. For the third transformation, the octuple mutant was selected for genetic manipulation using the CRISPR system. The selectable marker neo was inserted into a new endogenous locus, while the marker bar was removed without leaving a scar, and another gene was deleted without the use of a marker. This allows the marker bar to be reused in subsequent transformations.

Open in a separate window

Figure 7

CRISPR-Cas system constructed based on autonomously replicating plasmids with MAM1 sequence. Leveraging the instability of autonomously replicating plasmid with an AMA1 sequence, the plasmid will be lost through subculture without resistance and the strain with lost plasmid no longer has the resistance gene of plasmid. This allows the resistance gene to be reused in subsequent transformations. Red circles indicate the transformant containing hyg obtained by ATMT; blue circles indicate wild-type strains (control).

3.5. CRISPR-dCas system

Nuclease-deficient Cas9 (dCas9) is a protein produced by inserting the point mutations H841A and D10A into the nuclease domains of RuvC and HNH. As a consequence, a protein that was previously capable of cleaving DNA is now incapable of cutting DNA. This results in a protein that retains DNA binding activity but can no longer cleave DNA (Qi et al., 2013, 2021). dCas9 can be engineered as a programmable transcription repressor by preventing the binding of RNA polymerase (RNAP) to promoter sequences, or as a transcription terminator by blocking the running RNAP (Bikard et al., 2013). CRISPR-dCas9 systems can be used for CRISPR-mediated transcriptional activation (CRISPRa) or CRISPR-mediated transcriptional interference (CRISPRi) of target genes by fusing dCas9 with transcriptional activators or repressors, directly manipulating the transcription process without altering the DNA sequence (Cano-Rodriguez et al., 2016; Mahas et al., 2018). dCas9 can also regulate chromatin state and gene expression by fusing with different epigenetic modifiers, allowing for exploration of the interaction among the epigenome, regulatory elements, and gene expression (Keung et al., 2014; Kearns et al., 2015). Epigenetic modifiers have been fused to dCas9 and applied to mammalian cells, such as histone acetyltransferase (HAT) (Hilton et al., 2015), human lysine specific demethylase 1 (LSD1) (Kearns et al., 2015), disruptor of telomeric silencing 1-like (DOT1L) (Cano-Rodriguez et al., 2016), and DNA methyltransferases (Dnmt) (Vojta et al., 2016), have been fused to dCas9 and applied to mammalian cells. Additionally, the application of CRISPR-dCas9-mediated epigenetic modification systems to target gene regulation in Myxobacter (Peng et al., 2018), Malignant protozoa (Xiao et al., 2019), Arabidopsis thaliana (Lee et al., 2019). Under typical laboratory culture conditions, the poor expression of most biosynthetic gene clusters (BGCs) limits the entire metabolic potential stored in the genomes of fungi. The identification of bioactive secondary metabolites driven by genomics may be sped up by the CRISPRa of fungal BGCs. The CRISPRa of fungal BGCs could accelerate genomics-driven bioactive secondary metabolite discovery. In 2020, the first CRISPRa system for filamentous fungi was formed by fusing dCas12a with a transcription activator, VP64-p65-Rta (VPR). This system has been shown to successfully activate a fluorescent reporter in Aspergillus nidulans and can also target the native nonribosomal peptide synthetase-like (NRPS-like) gene micA in both chromosomal and episomal contexts, resulting in increased production of the compound microperfuranone. Furthermore, multigene CRISPRa led to the discovery of the mic cluster product, dehydromicroperfuranone (Roux et al., 2020). Li et al. (2021) successfully applied CRISPR/dCas9-based histone epigenetic modification systems in A. niger to regulate the expression of secondary metabolic genes breF, fuml, and fwnA.

3.6. Base editors

There is no denying the fact that CRISPR-Cas systems, which introduce DSB for gene editing at target genes, may have a negative impact on genome stability (Doudna and Charpentier, 2014). In contrast, base editors are considered to be more reliable and safer gene editing tools as they can efficiently and accurately edit target genes without generating DSBs (Figure 3E) (Lei et al., 2021). Base editors consist of a catalytically inactive form of Cas9 (dcas9) or a Cas9 nickase mutant (nCas9), cytidine deaminases, including apolipoprotein B mRNA-editing enzyme catalytic subunit 1 (APOBEC1) and activation-induced cytidine deaminase (AID), as well as a uracil glycosylase inhibitor (UG1) (Komor et al., 2016; Nishida et al., 2016). Depending on the type of single base substitution achieved by the deaminases, base editors can be divided into adenine base editors (ABEs) and cytosine base editors (CBEs). ABEs can directly convert A to G, while CBEs can convert C to T. For example, ABEs can disrupt gene transcription by converting the promoter codon ATG to ACG, while CBEs can also convert codons such as CAA, CAG, and CGA to termination codons TAA, TAG, and TGA, respectively, to prematurely terminate gene expression (Komor et al., 2016). These base editors can be used for gene inactivation, gene function research, and targeted modification of biological metabolism, and are gradually being applied to animals, plants, bacteria, yeast, and filamentous fungi. In 2019, scientists successfully applied this technology to gene editing in Aspergillus niger for the first time. They used a cytosine base editor fused with nCas9 and rat cytosine deaminase (CD) rAPOBEC1 to convert cytosine in the target sequence recognized by CRISPR-Cas into thymine, showing good editing efficiency in the genes pyrG, fwnA, and prtT (Huang et al., 2019). Zhang et al. (2022) developed three cytosine base editors (Mtevo-BE4max, MtGAM-BE4max, and Mtevo-CDA1) in the thermophilic fungus Myceliophthora thermophila and effectively silenced the amds, cre-1, and Mtclr-2 genes by accurately converting three codons (CAA, CAG, and CGA) to stop codons (TAA, TAG, or TGA). Among these editors, the Mtevo-CDA1 editor has an editing efficiency of up to 92.6%, making it a more suitable tool for cytosine base editing in Myceliophthora thermophila.