The filamentous fungus Aspergillus niger is an important cell factory in the biotech industry, which is used to produce organic acids such as citric and gluconic acid, as well as proteins, like glucoamylase or phytase [1, 2]. Genetic engineering offers a powerful approach to enhance filamentous fungi in terms of their productivity as well as to minimize undesirable traits like side-product formation. These engineering approaches started in fungi by transforming plasmid [3, 4] or cosmid DNA [5], which was introduced into the genome, relying on ectopic genome integration through non-homologous end joining (NHEJ). Additionally, homology-directed repair (HDR) using typically linear expression cassettes with homologous 5´and 3 ´ flanking regions were employed for genome deletions [6]. In Saccharomyces cerevisiae, DNA double-strand break repair is facilitated primarily via HR [7] and is mediated by the RAD52 epistasis group: RAD50, RAD51, RAD52, RAD54, RAD55, RAD57, RAD59, RFA1, MRE11, XRS2, and RDH54/TID1. This group of genes is highly conserved amongst eukaryotes, including A. niger [4, 8, 9]. However, the occurrence of HR events in A. niger was found to be low, standing at 1.78–7%, with NHEJ appearing to be the primary DNA repair mechanism [4, 9, 10]. The introduction of specific genetic modifications in A. niger is a challenging task due to this factor. Upon the deletion of NHEJ factors Ku70 (kusA in A. niger) and Ku80 (kusB in A. niger), the occurrence of HR significantly increased (65 ->80%) [9, 10]. Through this, site-specific gene editing became more accessible and NHEJ deficient strains are thus used by many groups as a tool to enable targeted gene engineering. However, it needs to be considered that NHEJ deficient strains have a slightly higher mutation rate than wild type strains which is relevant if a strain is frequently passaged or kept in a continuous culture [11].
A homologous transformation system for A. niger based on the pyrG gene was described by van Hartingsveldt et al., 1987 [12]. It was found that the transformation frequency based on the pyrG gene was at least tenfold higher than the heterologous transformation system for A. niger using the amdS gene and argB gene of Aspergillus nidulans, and the pyr4 gene of Neurospora crassa [12–14].
Nødvig et al., 2015 [15] were the first to apply CRISPR/Cas9 [16] to several species of Aspergillus including A. niger. Since then, the usage of this system has been adapted and improved, for example, by finding suitable promoters for guide RNA expression based on 5S RNA [17] or tRNA promoters [18] and the topic was well reviewed [19, 20].
In addition to CRISPR/Cas9 systems, alternative gene editing and integration tools for filamentous fungi are available. These include systems based on the Cre-loxP system involving site-specific recombination events [21–23] and the FLP/FRT system, which, similar to the Cre-loxP system, relies on site-specific recombination events [24]. Both strategies are recognized as efficient genetic engineering tools. However, the insertion of specific recombination sites into the genome is necessary and therefore, not scarless.
In 2015 we proposed a toolkit for metabolic pathway construction and genetic engineering in A. niger [25]. This system consists of a modular vector construction system called GoldenMOCS, based on the Golden Gate cloning approach [26], and a gene integration system for A. niger using CRISPR/Cas9 and self-replicating plasmids. The GoldenMOCS platform enables the versatile integration of host-specific parts such as promoters, terminators and resistance cassettes, replication origins or genome integration loci to customize the plasmid to the needs of the experiment and the host cell to be modified. Parts libraries for other organisms like Pichia pastoris and Yarrowia lipolytica are available [27, 28]. The fungal gene integration system uses a pyrG split-marker approach in combination with a transient Cas9 expression to enable selection on the integration event. For this approach, two plasmids are used that can be constructed with the GoldenMOCS pipeline: a Cas9/sgRNA-containing plasmid and a plasmid containing the integration cassette. The plasmids are co-transfomed into A. niger and can be transiently maintained in the fungal host using a size-reduced AMA1 version, which is readily lost after hygromycin selection is stopped. A special feature of the system is that the linear integration cassette can be released from the integration plasmid in vivo by Cas9 thereby the same gRNA/Cas9 complex is used to cut the plasmid and the genomic pyrG locus.
Upon successful homologous recombination of the pyrG split-marker, uridine prototrophy is restored, which is exploited as a selection marker for the integration event. Due to the modular character of the GoldenMOCS system up to eight different expression cassettes can be integrated into the pyrG locus using this strategy. The strains obtained in this way most likely have the cassette correctly integrated into the pyrG locus, with a minimal screening effort [29].
In this study, we evaluate this integration system and its HDR efficiency at the targeted pyrG locus of A. niger and can confirm the previously reported high targeting efficiency of the system. In addition, we observed a novel mixed-type repair mechanism in which the double-strand break mediated by CRISPR/Cas9 was simultaneously repaired by HDR and on the other side of the integration cassette by NHEJ.