We optimised the Agrobacterium-mediated transformation protocol of callus to establish a robust and effective genetic transformation system in F. tataricum. Previous studies have demonstrated the successful induction of morphogenic calli from immature embryos, characterised by high morphological and genetic stability, and regeneration potential, making it an ideal target for transformation (Betekhtin et al., 2017; Betekhtin et al., 2019). Transformation efficiency was evaluated using A. tumefaciens GV3101 carrying a GUS expression vector assembled using GreenGate parts (pGG-CaMV35S-GUS-kanR) ( Figure 1A ). GUS staining was performed on leaves of regenerated plants to assess the transformation efficiency. Out of the total 119 plants obtained from multiple transformation rounds, 24 plants exhibited blue precipitation indicating GUS enzyme activity, resulting in a transformation efficiency rate of 20.2% ( Figures 1B-D ; Table 1 ; Supplementary Figure S1 ). Some of the regenerated plants displayed mosaicism (8 plants, 6.7%), with only certain parts of the leaves showing successful transformation and GUS expression ( Figure 1C ). Additionally, we introduced a GFP expression vector (pGG-CaMV35S-GFP-NLS-kanR) carrying GFP with nucleolar localisation sequences (NLS) into the callus using Agrobacterium cells. Microscopic examination confirmed the efficient expression of GFP and its nuclear targeting in the callus and regenerated plants ( Figures 1F–H ).

The transformation of the morphogenic callus was performed using a callus in the middle of the passage, which exhibited visible proembryogenic cell complexes (PECCs) on its surface. The morphogenic callus is comprised of both ‘soft’ callus and PECCs. During the cyclic development of the callus, the mature PECCs undergo disintegration. PECCs contain subsurficial meristematic cells, the source of embryogenically determined cells from which new PECCs arise (Betekhtin et al., 2017). However, for the transformation process, the entire callus containing both PECCs and ‘soft’ callus cells was utilized. Typically, it takes approximately 4-6 weeks after transformation for the young PECCs to become visible on the callus surface ( Figure 1I ). Since the PECCs have the ability to regenerate into plants, only they are transferred to the regeneration medium. A detailed workflow outlining the steps of callus induction, transformation, and regeneration can be found in Figure 1J .

Despite the unavailability of genomic data for the k-17 sample of F. tataricum used in this study, we successfully designed primers based on the genomic sequence of F. tataricum cv. Pinku1 for amplifying a partial sequence of the PDS gene. The obtained sequence was utilised for designing a specific sgRNA targeting the first exon of the FtPDS gene ( Figure 2A ). The selected sgRNA sequence (5’-ACACAGCCCACTTCTACCCA-3’) exhibited high specificity with no off-target sites and demonstrated a predicted activity score of 0.305 and a specificity score of 100%. To assess the potential broader applicability of the designed sgRNA targeting the PDS gene, we aligned the partial PDS sequence obtained from the k-17 sample with the available genomic sequences of F. tataricum and F. esculentum. The alignment revealed high similarity in the coding region ( Figure 2B ). This indicates that the designed sgRNA can potentially be utilised for other F. tataricum cultivars and F. esculentum. The primers, incorporating the specific overlap sequence and FtPDS-targeting sgRNA, were assembled and introduced into the AarI-digested vector pGG-Cas9-GFP-AtU6-lacZ-kanR ( Figure 2C ). The resulting vector, pGG-Cas9-GFP-AtU6-FtPDS-kanR, was identified through blue-white screening, confirmed by sequencing, and introduced to Agrobacterium cell that were used for transformation. Out of 30 regenerated plants, six plants displayed an albino phenotype, indicating the potential inactivation of both copies of the FtPDS gene ( Figures 2D-F ). This corresponded to a mutation of both copies efficiency rate of 20% (Table 2). In one plant, we have observed mosaicism ( Figure 2G ). The PCR reaction amplifying the part of Cas9 gene sequence showed its presence in six albino plant, while being absent in the wild-type plants ( Figure 2I ). To confirm the inactivation of the FtPDS gene, we amplified the FtPDS gene region, cloned it into the pGEM vector, and performed sequencing analysis ( Figure 2J ). The original sequencing data are presented in the Supplementary Data 1 . The sequencing results revealed the presence of single or double nucleotide indels in the expected region targeted by the designed sgRNA. These introduced mutations caused a frame shift in the reading frame, resulting in a premature stop codon and subsequent inactivation of the PDS enzyme. The pds mutants showed diminished growth rates comparing to the wild-type ( Figures 2D-F, H ). Notably, the mutants displayed distinct pink pigmentation in their petioles and leaf veins ( Figures 2D-F ). This phenotypic change could be attributed to the accumulation of anthocyanins, colourful compounds abundantly found in Fagopyrum species (Fang et al., 2019). In the wild-type, the pink colour can be observed in the stalks of the plants ( Figure 2H ) but not in the petioles and leaf veins, possibly due to the presence of obscuring anthocyanins green chlorophyll (Saniewski et al., 2022).

The study utilised the seeds of F. tataricum (sample k-17) obtained from the N. I. Vavilov Institute of Plant Genetic Resources collections in Saint Petersburg, Russia. F. tataricum sample k-17 represents a commonly cultivated landrace of F. tataricum, and seeds can be obtained upon request from the publication’s authors. The plants were grown under field conditions from May to September. The morphogenic calli lines were derived from immature zygotic embryos in the dark at 26 ± 1°C using an RX medium as described previously (Betekhtin et al., 2017; Betekhtin et al., 2019; Zaranek et al., 2023a). The RX medium consisted of B5 mineral salts supplemented with 2.0 mg/L thiamine-HCl, 1.0 mg/L pyridoxine-HCl, 1.0 mg/L nicotinic acid, 2000 mg/L casein hydrolysate, 2.0 mg/L 2.4-D, 0.5 mg/L indolylacetic acid, 0.5 mg/L naphthylacetic acid, 0.2 mg/L kinetin, 2.5% sucrose and 0.8% agar (Gamborg et al., 1968). The morphogenic calli cultures were subcultured every four weeks. A callus in the middle of the passage containing visible PECCs on the surface was used for the morphogenic callus transformation.

The partial region of the PDS gene in F. tataricum (FtPinG0302920600.01) was amplified using primers designed based on the F. tataricum cv. Pinku1 genome assembly as described by Zhang et al. (2017) ( Table S1 ). The amplified PCR product was subsequently cloned into the pGEM-T Easy vector (Promega, USA) and sequenced. The partial sequence of FtPDS data have been submitted to the GenBank databases under accession number OR334360. To identify target sequences within the FtPDS gene, the Find CRISPR Sites tool in Geneious Prime 2022.0.2 was utilised. The activity of the target sequences was assessed using Doench et al. (2014) scoring method, while specificity was evaluated following the Hsu et al. (2013) criteria. For the alignment of the first exon sequence of the PDS gene, the genomes of F. tataricum cv. Pinku1, F. esculentum cv. DASHA, and the sequencing results for F. tataricum sample k-17 were utilized (Zhang et al., 2017; Penin et al., 2021).

The vectors for GFP (pGG-CaMV35S-GFP-NLS-kanR) and GUS (pGG-CaMV35S-GUS-kanR) expression were generated using the GreenGate system as described by Lampropoulos et al. (2013). The plasmid kit (GreenGate system) used to generate plant transformation constructs was a gift from Jan Lohmann (Addgene kit # 1000000036). The vectors were assembled using the Golden Gate reaction by ligating vectors: pGGZ004 (Lupanga et al., 2020) (vector backbone conferring spectinomycin resistance), pGGA004 (CaMV35S promoter), pGGB005 (N-tag, B-dummy), pGGD002 (C-tag, B-dummy), pGGE001 (RBCS terminator), pGGF007 (pNOS : KanR:tNOS), and pGGC012 (CDS, GFP with NLS) for GFP expression and pGGC051 (CDS, GUS) for GUS expression. The construction of the pGG-Cas9-GFP-AtU6-lacZ-kanR vector for sgRNA cloning involved several steps. The primers utilised for amplifying the elements required for vector and sgRNA construction are provided in Table S1 . Firstly, the plant codon-optimised Cas9 gene was amplified from the pYPQ167 vector, and the STOP codon was removed during amplification (Lowder et al., 2015). The amplified Cas9 gene sequence was then cloned into the empty CDS entry vector (pGGC000), resulting in the vector pGG-Cas9(ΔSTOP). For sgRNA cloning, the pGG-F-AtU6-BbsI-ccdB-BbsI-G vector (contains promoter AtU6-26) was modified by removing the ccdB cassette through digestion with the BbsI restriction enzyme (Decaestecker et al., 2019). The resulting linear backbone was ligated with the AarI-lacZ-AarI PCR product, which was digested with the AarI enzyme, creating the pGG-F-AtU6-AarI-lacZ-AarI-G vector. The AarI-lacZ-AarI insert was cloned from vector JD633 (Debernardi et al., 2020). JD633 was a gift from Jorge Dubcovsky (Addgene plasmid # 160393; http://n2t.net/addgene:160393; RRID : Addgene_160393). Finally, the final vector, pGG-Cas9-GFP-AtU6-lacZ-kanR, was assembled by ligating vectors: pGGA004 (CaMV35S promoter), pGGB005 (N-tag, B-dummy), pGG-Cas9(ΔSTOP), pGGD001 (linker-GFP), pGGE001 (RBCS terminator), pGG-F-AtU6-AarI-lacZ-AarI-G (empty vector for cloning sgRNA), and pGGK-AG (vector backbone). The oligos containing the sgRNA sequence and 5’ overlap sequence (5’-ATTG-N20-3’ and 5’-AAAC-N20 (reverse complement)-3’) specific to FtPDS were annealed and cloned into the empty gRNA destination vector, pGG-Cas9-GFP-AtU6-lacZ-kanR. This resulted in the generation of the vector pGG-Cas9-GFP-AtU6-FtPDS-kanR. The proper assembly of the donor and final vectors was verified by Sanger sequencing. Subsequently, the vectors pGG-Cas9-GFP-AtU6-FtPDS-kanR, pGG-CaMV35S-GFP-NLS-kanR and pGG-CaMV35S-GUS-kanR were transformed into A. tumefaciens GV3101 electrocompetent cells (Gold Biotechnology) and selected on LB plates supplemented with rifampicin (25 μg/mL), spectinomycin (25 μg/mL), and gentamycin (50 μg/mL), as previously described (Hus et al., 2020). For all PCR reactions Q5 High-Fidelity DNA Polymerase was used (New England Biolabs, USA). The detailed schematics and sequences of generated vectors are presented in Supplementary Figure S2 and Supplementary Data 2 .

To initiate the Agrobacterium culture, 5 μL of a glycerol stock of A. tumefaciens GV3101 transformed with the respective vectors was inoculated into 1 mL of LB medium containing rifampicin (25 μg/mL), spectinomycin (25 μg/mL), and gentamycin (50 μg/mL). The culture was then incubated at 28°C with shaking at 200 rpm for 24 hours. For plating the overnight Agrobacterium culture, 200 μL of the culture was spread onto LB agar plates supplemented with spectinomycin (25 μg/mL), gentamycin (50 μg/mL), and 30 mg/L acetosyringone. For the transformation process, the Agrobacterium cells were scraped from the plates and added to liquid RX medium supplemented with 30 mg/L acetosyringone and 0.1% Pluronic F-68. The OD₆₀₀ of the culture was adjusted to 0.8, and the culture was incubated at 28°C with shaking at 220 rpm for 1 hour before the callus transformation.

Approximately 5 g of the morphogenic callus were placed in glass Petri dishes (diameter 100 mm). The calli were then inoculated with 30 mL of an Agrobacterium suspension and incubated for 10 minutes. Subsequently, the Agrobacterium liquid was removed. Then callus was transferred on two filter discs (diameter 110 mm) followed by a 15-min desiccation treatment to limit subsequent overgrowth of Agrobacterium and to promote transformation through an osmotic effect. Next, the infected callus was transferred onto two filter discs (diameter 110 mm) placed on 2 ml of liquid RX medium supplemented with 30 mg/L acetosyringone. The plates were sealed with Parafilm and kept in the dark at 26°C for three days. Following the co-cultivation period, the callus was transferred to RX medium containing 50 mg/L kanamycin, 300 mg/L timentine, and 200 mg/L cefotaxime (all antibiotics from Duchefa) and incubated in the dark at 26 ± 1°C for three weeks. After three weeks, the callus was transferred to a fresh medium with the same antibiotic concentrations and cultivated for three more weeks.

After six weeks on the medium with antibiotics, only newly emerged PECCs were isolated and transferred onto the regeneration medium. The regeneration medium consisted of MS basal salts supplemented with 50 mg/L kanamycin, 300 mg/L timentine, 200 mg/L cefotaxime, 2.0 mg/L benzylaminopurine, 1.0 mg/L kinetin, 30 g/L sucrose, and 3 g/L gelrite. The cultures were maintained in a growth chamber at 28 ± 1°C under a 16/8 h (light/dark) photoperiod, with a light intensity of 80 µmol m-2 s-1. Subculturing of calli with developing somatic embryos or shoots was performed every three to four weeks. Fully developed shoots were further transferred to an MS medium containing vitamins but without plant growth regulators, supplemented with 25 mg/L kanamycin, 300 mg/L timentine, 200 mg/L cefotaxime, 30 g/L sucrose, and 3 g/L gelrite.

GUS activity in the calli and leaves of F. tataricum was monitored throughout the transformation experiments. The GUS assay was performed by incubating the calli and leaves in a histochemical buffer containing 0.1 M sodium phosphate buffer (pH 7.0), 50 mM EDTA, 0.5 mM K₃Fe(CN)₆, 0.5 mM K₄Fe(CN)₆, 0.1% Triton-X-100, and 1 mg/mL X-gluc (5-bromo-4-chloro-3-indolyl-β-glucuronidase) overnight at 37°C (Jefferson et al., 1987). After incubation, the calli and leaves were washed with deionised water and cleared with 70% ethanol.

To assess the presence of the GFP signal, a portion of the fully developed leaf was transferred onto a slide in a drop of distilled water. The sample was then covered with a coverslip. The GFP signal was examined using a Zeiss Axio Imager Z2 microscope with an AxioCam Mrm monochromatic camera (Zeiss). The images were captured using the corresponding software and narrow-band filters designed for GFP visualisation. The bright field images were captured using a KEYENCE VHX-7000 digital microscope (Keyence, Japan).

The DNA from the leaves of regenerated plants were isolated using the C-TAB isolation protocol as described previously (Hus et al., 2020). The partial sequence of the FtPDS gene was amplified from the isolated DNA samples and cloned into the pGEM-T Easy vector. Four to six bacterial colonies were randomly selected for each plant and subjected to sequencing analysis to ensure coverage of both gene copies. The obtained sequencing data were aligned with the reference sequence using Geneious Prime 2022.0.2 software, allowing for the identification of any mutations in the FtPDS gene.