Gene function studies in Aspergillus niger

This thesis aimed to reassess the practice of (developmental) mutant analysis in the understudied fungus A. niger, using a novel mutagenesis system based on CRISPR-Cas9. To this effect, the optimization of Aspergillus gene editing plasmids, towards the cloning of gRNA libraries of potentially any size, is described in Chapter 2. A gRNA library of small size, targeting five independent markers, was pooled and transformed into A. niger to determine proportions of combinatorial mutants. Although single and double mutants were the most common, all 36 possible single and combination mutants were observed. When a pool of gRNAs was targeted to 11 developmental genes, a combination phenotype was attributed to the inactivation of aslB and flbD. Inactiavion of aslB was shown to supress the delayed conidiation phenotype of flbD inactivation. Chapter 3 aimed to scale up A. niger pooled libraries to a genome-wide size. A total of 9,635 gRNAs were selected, targeting each a single gene from the 11,846 genes encoded in the genome. The library pool was successfully introduced in A. niger. However, the auxotrophic pyrG transformant recovered by FOA selection carried a spurious gRNA sequence. Therefore, a strategy was devised for co-transforming libraries with another selectable gRNA plasmid. Using high ratios of this plasmid, the transformants could be stringently co-selected for limited uptake of library plasmids, which permitted the isolation and identification of a development mutant that mapped to the A. nidulans nsdC homologue. Chapter 4 describes the phenotypes of the A. niger developmental genes pkaC, brlA, and flbA when inactivated individually or in combination. A role of BrlA and PkaC as negative interacting regulators of mycelium density and sclerotia formation was proposed. Also, inactivated pkaC phenotypes were affected by the absence of FlbA but not if brlA was also inactivated, demonstrating the need of multiple gene mutations to fully understand the Aspergillus developmental program. Furthermore, the origin of the intense yellow pigmentation in the flbA- mutant was investigated in a suppressor screen after mutagenesis with a genome-wide gRNA library pool, which revealed a gene involved in yellow pigmentation, as well as three other genes that may be involved as well in production of this pigment. Transformation experiments with the genome-wide library requires vast supplies of protoplasts. Thus, Chapter 5 addressed the problem of unreliable commercial Trichoderma harzianum lysing enzymes, characterized by market shortage, high cost and low efficiency. The solution consisted in growing T. harzianum on A. niger cell walls, resulting in a culture medium with high chitinase activity being more effective for protoplasting A. niger than the commercial enzymes. The transcriptomic response of T. harzianum was analyzed and candidate genes involved in cell wall lysis were highlighted.