A comparison of three different delivery methods for achieving CRISPR/Cas9 mediated genome editing in Cichorium intybus L.

Salvagnin, Umberto; Unkel, Katharina; Sprink, Thorben; Bundock, Paul; Sevenier, Robert; Bogdanović, Milica; Todorović, Slađana; Cankar, Katarina; Hakkert, Johanna Christina; Schijlen, Elio; Nieuwenhuis, Ronald; Hingsamer, Maria; Kulmer, Veronika; Kernitzkyi, Michael; Bosch, Dirk; Martens, Stefan; Malnoy, Mickael


Root chicory (Cichorium intybus L. var. sativum) is used to extract inulin, a fructose polymer used as a natural sweetener and prebiotic. However, bitter tasting sesquiterpene lactones, giving chicory its known flavour, need to be removed during inulin extraction. To avoid this extraction and associated costs, recently chicory variants with a lower sesquiterpene lactone content were created by inactivating the four copies of the germacrene A synthase gene (CiGAS-S1, -S2, -S3, -L) which encode the enzyme initiating bitter sesquiterpene lactone biosynthesis in chicory. In this study, different delivery methods for CRISPR/Cas9 reagents have been compared regarding their efficiency to induce mutations in the CiGAS genes, the frequency of off-target mutations as well as their environmental and economic impacts. CRISPR/Cas9 reagents were delivered by Agrobacterium-mediated stable transformation or transient delivery by plasmid or preassembled ribonucleic complexes (RNPs) using the same sgRNA. All methods used lead to a high number of INDEL mutations within the CiGAS-S1 and CiGAS-S2 genes, which match the used sgRNA perfectly; additionally, the CiGAS-S3 and CiGAS-L genes, which have a single mismatch with the sgRNA, were mutated but with a lower mutation efficiency. While using both RNPs and plasmids delivery resulted in biallelic, heterozygous or homozygous mutations, plasmid delivery resulted in 30% of unwanted integration of plasmid fragments in the genome. Plants transformed via Agrobacteria often showed chimerism and a mixture of CiGAS genotypes. This genetic mosaic becomes more diverse when plants were grown over a prolonged period. While the genotype of the on-targets varied between the transient and stable delivery methods, no off-target activity in six identified potential off-targets with two to four mismatches was found. The environmental impacts (greenhouse gas (GHG) emissions and primary energy demand) of the methods are highly dependent on their individual electricity demand. From an economic view - like for most research and development activities - employment and value-added multiplier effects are high; particularly when compared to industrial or manufacturing processes. Considering all aspects, we conclude that using RNPs is the most suitable method for genome editing in chicory since it led to a high efficiency of editing, no off-target mutations, non-transgenic plants with no risk of unwanted integration of plasmid DNA and without needed segregation of transgenes.