DNA-free CRISPR-Cas9 gene editing of tetraploid tomatoes using protoplast regeneration

Plant materials

Sterile S. peruvianum plantlets were propagated by cutting and growing them in half-strength Murashige and Skoog (1/2 MS) medium supplemented with 30 mg/L sucrose and 1% agar, pH 5.7. The plantlets were incubated in a 26°C culture room (12 h light/12 h in dark, light intensity of 75 µmol m−2 s−1). The plantlets were cut and subcultured in fresh medium monthly.

Protoplast isolation and transfection

Protoplast isolation and transfection of S. peruvianum were performed following our previously published method with minor modifications (Hsu et al., 2019). Protoplasts were isolated from the stems and petioles of _in vitro-_grown plantlets. Five or more stems (approximately 5 cm/each, total 0.2-0.25 g) were used to isolate roughly 1 × 106 protoplasts. These materials were place in a 6-cm glass Petri dish with 10 mL digestion solution 1/4 Murashige and Skoog (MS) liquid medium containing 1% cellulose and 0.5% macerozyme, 3% sucrose, and 0.4 M mannitol, pH 5.7 and cut into 0.5-cm-wide strips longitudinally. The material was incubated at room temperature in the dark overnight. The digested solution was diluted in 10 mL W5 (154 mM NaCl, 125 mM CaCl2, 5 mM KCl, 2 mM MES, and 5 mM glucose) solution and filtered through a 40-µm nylon mesh. The sample was centrifuged at low speed (360 × g) for 3 min to collect the protoplasts. The protoplasts were purified in 20% sucrose solution and washed three times with W5 solution. The protoplasts were transferred to transfection buffer (1/2 MS medium supplemented with 3% sucrose, 0.4 M mannitol, 1 mg/L NAA, and 0.3 mg/L kinetin, pH 5.7) and adjusted to a concentration of 3 × 105 cells/mL.

The protoplasts were transfected with plasmids by PEG-mediated transfection (Woo et al., 2015; Lin et al., 2018). A 400-µL sample (1.2 × 105 protoplasts) was combined with 40 µL of CRISPR reagent (DNA: 20-40 µg; RNP: 10 µg) and mixed carefully. The same volume (440 µL) of PEG solution was added to the sample, mixed, and incubated for 30 min. To end the reaction, 3 mL of W5 was added, and the sample was mixed well. Transfected protoplasts were collected by centrifugation at 360 × g for 3 min. The protoplasts were washed with 3 mL of W5 and centrifuged at 360 × g for 3 min. The target sites are shown in Table 1.

CRISPR/Cas reagents

The SpCas9 vector for dicot transformation (pYLCRISPR/Cas9P35S-N) (Ma et al., 2015) was isolated using a Plasmid Midi-prep kit (Bio-Genesis). Preparation of Cas9 protein and sgRNA and Cas9 RNP nucleofection were performed according to Huang et al., 2020. Cas9 RNP complexes were assembled immediately before nucleofection by mixing equal volumes of 40 μM Cas9 protein and 88.3 μM sgRNA at a molar ratio of 1:2.2 and incubating at 37°C for 10 min.

Protoplast regeneration

Pooled protoplast DNA was used as a template to amplify the target genes for validation by sequencing. The putatively edited protoplasts were transferred to 5-cm-diameter Petri dishes containing 3 mL 1/2 MS liquid medium supplemented with 3% sucrose, 0.4 M mannitol, 1 mg/L NAA, and 0.3 mg/L kinetin for plant regeneration. Calli formed from the protoplasts after 1 month of incubation in the dark. The calli were subcultured in a 9-cm-diameter Petri dish containing fresh medium with cytokinin for 3-4 weeks in the light. Calli that had turned green were transferred to solid medium containing the same plant growth regulators. The explants were subcultured every 4 weeks until shoots formed after several subcultures. The shoots were subcultured in solid rooting medium (HB1: 3 g/L Hyponex No. 1, 2 g/L tryptone, 20 g/L sucrose, 1 g/L activated charcoal, and 10 g/L Agar, pH 5.2) for adventitious roots formation.

Analysis of the genotypes of regenerated plants

Two pairs of primers were designed to amplify the sgRNA-targeted DNA region for each target gene. The PCR conditions were 94°C for 5 min, 35 cycles of denaturing (94°C for 30 s), annealing (55°C for 30 s), and polymerization (72°C for 30 s), followed by an extension reaction at 72°C for 3 min. The PCR product was sequenced by Sanger sequencing to confirm mutagenesis. The multiple sequences derived from mutated regenerated plants were bioinformatically separated using Poly Peak Parser (; (Hill et al., 2014)) or further confirmed by sequential T/A cloning and sequencing. The primer sequences are listed in Table S7.

Estimation of genome size

Fresh leaves were finely chopped with a new razor blade in 250 µL isolation buffer (200 mM Tris, 4 mM MgCl2-6H2O, and 0.5% Triton X-100) and mixed well (Dolezel et al., 2007). The mixture was filtered through a 40-μm nylon mesh, and the filtered suspensions were incubated with a DNA fluorochrome (50 μg/mL propidium iodide containing RNase A). The samples were analyzed using a MoFlo XDP Cell Sorter (Beckman Coulter Life Science) and an Attune NxT Flow Cytometer (Thermo Fisher Scientific). Chicken erythrocyte (BioSure) was used as an internal reference.

Whole genome sequencing

Leaves of S. peruvianum regenerates were harvested and genomic DNA was extracted using two independent protocols. A nuclei isolation protocol (Sikorskaite et al., 2013) was used on the wild type (SpB) sample to recover higher quality and quantity of DNA samples. Briefly, nuclei were extracted by 36mM sodium bisulfite, 0.35M Sorbitol, 0.1M Tris-base, 5mM EDTA, 2M NaCl, 2% (w/v) CTAB, and 2 ml 5% N-lauroylsarcosine sodium salt. The genomic DNA was then extracted by chloroform-isoamyl alcohol (24:1), ethanol precipitation, and further cleaned up by DNeasy Blood & Tissue Kit (69504, Qiagen) and AMPure (Beckman Coulter). The other nine samples used the chloroform-isoamyl alcohol (24:1) for DNA extraction, followed with Zymo Genomic DNA Clean & Concentrator-25 (D4064, Zymo), and Zymo OneStep PCR Inhibitor Removal Kit (D6030, Zymo) to obtain high quality genomic DNA. DNA integrity was checked using the D1000 Screen Tape on the Agilent TapeStation 4150 System with DIN value > 8. Genomic DNA were sheared using a Covaris E220 sonicator (Covaris) and paired-end sequencing libraries were constructed by the NEBNext Ultra DNA Library Prep Kit II for Illumina (E7370S, NEB). DNA libraries were validated again on the Agilent TapeStation 4150, and were quantified by qPCR (E7630, NEB). The 2×150 bp paired-end sequencing with average insert size of 700 bp was performed by Welgene Biotech on an Illumina NovaSeq 6000 platform.

WGS data analysis

Since there was no assembled S. peruvianum genome, high quality Illumina reads were mapped to the S. lycopersicum Heinz 1706 reference genome (SL4.0) (Hosmani et al., 2019) by the GPU-based NVIDIA Clara Parabricks package (NVIDIA). To determine the variant frequency, we used the deep learning-based Google DeepVariant (Yun et al., 2021) with ‘WGS model’ to identify variants. All samples were then combined by GLnexus (Yun et al., 2021) to perform ‘joint genotype calling’ using ‘DeepVariant’ model to combine samples. We then calculated the heterozygous allele frequency by dividing the read depth of the heterozygous allele (labeled as 0/1 by GLnexus) over the total read depth of the variant. A large chromosomal region with heterozygous allele frequency lower than 0.5 indicated either the chromosome region with low recombination rate or deletion of the chromosome fragments. To determine CNVs between samples, we used the cn.mops pipeline (Klambauer et al., 2012) to analyze mapped Illumina reads. To minimize the effects of repetitive sequence regions, we set the segment size to 3,000 bp and minimum number of segments as 10 to identify high confidence CNVs.

Quantitative real-time PCR (RT-qPCR)

Expression of four genes was analysed using real-time PCR. These genes were: SpSGS3, SpARF3, SpARF4, and SpRDR6. Transcripts of all four genes were profiled with three biological replications and each with at least three technical replications using the RNA samples of regenerants. RT-qPCR was carried out in 96-well optical reaction plates using the iQ™ SYBR® Green Supermix (Bio-Rad). The reference gene Actin and gene-specific primers for the RT-qPCR are listed in Supplementary Table S8.

Article TitleDNA-free CRISPR-Cas9 gene editing of tetraploid tomatoes using protoplast regeneration


Wild tomatoes are important genomic resources for tomato research and breeding. Development of a foreign DNA-free CRISPR-Cas delivery system has potential to mitigate public concern about genetically modified organisms. Here, we established a DNA-free protoplast regeneration and CRISPR-Cas9 genome editing system for Solanum peruvianum, an important resource for tomato introgression breeding. We generated mutants for genes involved in small interfering RNAs (siRNA) biogenesis, RNA-DEPENDENT RNA POLYMERASE 6 (SpRDR6) and SUPPRESSOR OF GENE SILENCING 3 (SpSGS3); pathogen-related peptide precursors, PATHOGENESIS-RELATED PROTEIN-1 (SpPR-1) and PROSYSTEMIN (SpProsys); and fungal resistance (MILDEW RESISTANT LOCUS O, SpMlo1) using diploid or tetraploid protoplasts derived from in vitro-grown shoots. The ploidy level of these regenerants was not affected by PEG-calcium-mediated transfection, CRISPR reagents, or the target genes. By karyotyping and whole genome sequencing analysis, we confirmed that CRISPR-Cas9 editing did not introduce chromosomal changes or unintended genome editing sites. All mutated genes in both diploid and tetraploid regenerants were heritable in the next generation. spsgs3 null T0 regenerants and sprdr6 null T1 progeny had wiry, sterile phenotypes in both diploid and tetraploid lines. The sterility of the spsgs3 null mutant was partially rescued, and fruits were obtained by grafting to wild-type stock and pollination with wild-type pollen. The resulting seeds contained the mutated alleles. Tomato yellow leaf curl virus proliferated at higher levels in spsgs3 and sprdr6 mutants than in the wild type. Therefore, this protoplast regeneration technique should greatly facilitate tomato polyploidization and enable the use of CRISPR-Cas for S. peruvianum domestication and tomato breeding.

One-sentence summary DNA-free CRISPR-Cas9 genome editing in wild tomatoes creates stable and inheritable diploid and tetraploid regenerants.

Login or Signup to leave a comment
Find your community. Ask questions. Science is better when we troubleshoot together.
Find your community. Ask questions. Science is better when we troubleshoot together.

Have a question?

Contact or check out our support page.