Escherichia coli (E. coli) DH5α [F⁻endA glnV44 thi-1 recA1 relA1 gyrA96 deoR nupG Φ80dlacZΔM15Δ(lacZYA-argF)U169, hsdR17(rK−mK+), λ⁻], A. tumefaciens (GV2260) [C58 background, rifampicin-resistant with the Ti plasmid (pTiB6s3)], and Escherichia coli (E. coli) helper P600 (Traore and Zhao, 2011).

N. benthamiana (PI 555478) plants were propagated in a growth chamber programmed for 16 h light (140 μmol m⁻²s⁻¹ cool white fluorescent irradiance) at 28°C and 8 h dark at 24°C Agrobacterium-mediated transient assays were conducted on three- to 4-week-old plants.

A 780 bp DNA fragment with deletion of the N and C-terminus of the recA gene was amplified from the genomic DNA of GV2260 using primers: recA For, 5′-caccatcgatcatgaagctcggt-3′ and recA Rev, 5′-gcgccggacttctcgacgat-3′. The PCR reaction as performed using the iProof™ high fidelity Taq DNA polymerase (Bio-Rad, Hercules, CA). The PCR program consisted of 1 cycle at 98°C (2 min), followed by 30 cycles at 98°C (30 s), 55°C (45 s), and 72°C (1 min), and finished with a 1 cycle extension at 72°C (7 min). The PCR product was separated on a 0.8% agarose gel, stained with 0.01% ethidium bromide solution, and visualized using the Gel-Document Image System™ under UV light (Bio-Rad).

The PCR product was purified using the AccuPrep™ Gel Purification Kit (Bioneer, Alameda, CA) and cloned into the TopoEntr/D™ vector (Invitrogen, Carlsbad, CA) following the instructions of the user manual. The derived plasmid vector was designated as TopoEntr-recA and has been sequenced at the core facility of the Virginia Bioinformatics Institute (Blacksburg, VA).

The suicide vector pLVC18L (Zhao et al., 2011) was modified by insertion of the SacB/R and the ccdB gene cassettes. The Npt2 promoter-SacB/R fragment was amplified through overlap PCR from pEG101-SacB/R and pDSK519-GFP (Matthysse et al., 1996; Traore and Zhao, 2011) using primers: 1846pLvc18 XbaNpt2 Infusion For1, 5′-TGC CATTGCTGCAGGTCGACTCTAGAGATATCACATGGCGATAGCTAGACT G-3′; 1777Npt2Pro_SacB/R Rv Rev1, 5′-GTG ATGGGTTAAAAAGGATCGATCCGCGCCATCAGATCC TTG-3′; 1778Npt2Pro_SacB/R Rv For2, 5′-CAA GGATCTGATGGCGCGGATCGATCCTTTTTAACCCAT CAC-3′; 1847pLvc18 XbaSacB Infusion Rev2, 5′-CTC GGTACCCGGGGATCCTCTAGAGATATCTTATTTGTTAACTGTTAATTG TCCT-3′.

The PCR product was cloned into the XhoI site of pLVC18L using a Gibson cloning kit (New England BioLabs Inc., Ipswich, MA). The derived construct was designated as pLVC18L-Npt2-SacB/R. A ccdB gene cassette (frame B) (Invitrogen) was further cloned into the SmaI site of pLVC18L-Npt2-SacB/R to generate pLVC18L-Npt2-SacB/R-DesB. The recA gene fragment from TopoEntr-recA was subcloned into pLVC18L-Npt2-SacB/R-DesB using a Gateway®; LR cloning kit (Invitrogen) following the instructions of the user manual. The derived plasmid construct was named pLVC18L-Npt2-SacB/R-recA, and has been confirmed by sequencing at the core facility of the Virginia Bioinformatics Institute (Blacksburg, VA).

The suicide vector pLVC18L-Npt2-SacB/R-recA was integrated to the genome of A. tumefaciens strain GV2260 by tri-parental conjugation and was selected on LB medium supplemented with tetracycline (10 μg/mL) as previously described (Traore and Zhao, 2011). A mutant GV2260 strain carrying the Npt2-SacB/R cassette was confirmed by PCR amplification of the tetracycline resistance gene and SacB/R gene using primers: tetracycline For, 5′-atgaaatctaacaatgcg ctcat-3′; tetracycline Rev, 5′-tacgagttgcatgataaagaa gaca-3′, and SacB-SacR For, 5′-cagcatatcatggcgtgt aatatg-3′; SacB-SacR Rev, 5′-ctcggtacccggggatcctctagagat atcttatttgttaactgttaattgtcct-3′. The derived mutant strain was designated as GV2260-SacB/R.

The YFP gene open reading frame plus the HA epitope tag was amplified from vector pEarleygate101 (Earley et al., 2006) with primers 2702pEG101-yfpHA For, 5′-ATTTGGAGAGGACACGctcgagAtgAGCAAGGGCGAGGAGCTGTTC ACCG-3′; 2703pEG101-yfpHA Rev, 5′-TCGACTGCAGAATTCGAAGCTTGAGctcgagATCTGAG-3′. The PCR product was gel purified and cloned into pEarleygate101 that had been digested with XhoI. The derived construct was designated as pEarleygate101-YFP-HA.

A putative tobacco WRKY transcription factor NbWRKY70 was identified from GenBank (accession number AF421157). To knock out NbWRKY70, we identified a guiding RNA (GCAATCGACGGGTTAATTCGCGG) targeting the NbWRKY70 gene. An Arabidopsis U6 promoter, NbWRKY70 guiding RNA, and the PAM terminator were amplified through overlap-PCR using primers 2284AtU6gRNA common For1, CAGCAACTCATTACAACTTGTTTaagctttcgttgaacaacgga; 2285AtU6gRNA common Rev1, CGACTCTAGACACGGGGTGGTTTaaaaaaagcaccgactcggtgcc; 2846NbWRKY70 gRNA For, GCAATCGACGGGTTAATTCGgttttagagctagaaatag; 2847NbWRKY70 gRNA Rev, CGAATTAACCCGTCGATTGCaatcactacttcgactcta.

The PCR product was gel purified and cloned into the PmeI site of pM3UT-Cas9 using a Gibson cloning kit (New England BioLabs), where it contains a Cas9 gene driven by the Arabidopsis Ubiquitin 10 promoter. The Cas9 gene was originally codon optimized and synthesized based on the Arabidopsis genes (Zachary Nimchuk, unpublished data). The derived construct was designated as pgRNA-NbWRKY70.

The plasmid vectors pEarleygate101-YFP-HA and pgRNA-NbWRKY70 were conjugated into A. tumefaciens strain GVV2260 and GV2260-SacB/R using tri-parental conjugation, selected on LB medium supplemented with rifampin 100 μg/mL, and Kanamycin 50 μg/mL or Spectinomycin 50 μg/mL as previously described (Traore and Zhao, 2011).

Agrobacterium-mediated transient assays in N. benthamiana plants were performed as described previously (Wydro et al., 2006). In brief, the Agrobacterium strains were streaked on Yeast Extract Tryptone (YT) media supplemented with rifampicin 100 μg/mL, tetracycline 10 μg/mL, and kanamycin 50 μg/mL and incubated at 28°C for 2 days. Bacterial cells were harvested and re-suspended in induction buffer composed of 10 mM MgCl₂, 10 mM MES (pH 5.6), and 100 μM acetosyringone and incubated for 3 h at room temperature. The bacterial inoculums were adjusted to OD₆₀₀ nm = 0.6 and infiltrated into the stomata of the fully expanded N. benthamiana leaves using a 1-mL blunt-end syringe without a needle. The inoculated plants were incubated at room temperature under continuous light for 20–48 h before the detection of expressed proteins. The fluorescent signal of YFP-HA fusion protein was monitored 24 h after inoculation by fluorescent microscopy (Zeiss Axio Observer.A1, Carl Zeiss MicroImaging, Inc., Thornwood, NY).

Agrobacterium strain GV2260 or GV2260-SacB/R carrying plasmid pgRNA-NbWRKY70 were used for tobacco transformation following a previously described protocol (Horsch et al., 1989). In brief, the fully expanded leaf from a 4-week-old N. benthamiana plant was collected and sterilized in 10% bleach for 20 min. The leaf was cut into 1 cm² leaf disks that were infected with Agrobacterium culture diluted to OD₆₀₀ = 0.1. The infected leaf disks were co-cultured on MS medium supplemented with 6-BA (1 mg/L) and NAA (0.1 mg/L) and 3% maltose at 25°C in the dark for 2 days. The infected leaf disks were soaked in liquid MS medium for 5 min and then rinsed one time with liquid MS medium. The leaf disks were then blotted dry and transferred to a selection medium (MS medium supplemented with kanamycin 300 mg/L, cefotaxime 150 mg/L and 3 or 5% sucrose). Each treatment has at least 100 leaf disks with three replicates.

The selection mediums were incubated at 25°C under continuous light for 25–30 days for shoot regeneration. The transgenic shoots were transferred to rooting medium (MS medium supplemented with Kanamycin 100 mg/L and 3% sucrose).

The putative transgenic tobacco plants were confirmed by PCR with primers: PMOA36-TBS PmeI For, 5′-TGATAGAGTAGTTCATAT GGA-3′ and PMOA36-TBS PmeI Rev 5′-GCTTCC CAACCTTACCAGAG-3′. The PCR products were gel purified and sequenced at the core facility at Virginia Bioinformatics Institute.

Bacterial genomic DNAs were isolated using a ZR Fungal/Bacterial DNA MiniPrep™ (Zymo Research Corporation, Irvine, CA). Plasmid DNAs were isolated using an AccuPrep™ Plasmid Extraction Kit (Bioneer Corporation, Alameda, CA). Plant genomic DNAs were isolated by using the CTAB method as previously described (Zhang et al., 2013).

To generate a mutant Agrobacterium strain that is sensitive to sucrose, a suicide vector pLVC18L carrying the SacB-SacR gene cassette was integrated into the recA gene locus in the genome of A. tumefaciens strain GV2260 through marker-exchange mutagenesis. The mutant strain was named GV2260-SacB/R. The integration of the SacB-SacR gene cassette in GV2260-SacB/R was confirmed by PCR amplification. As shown in Figure 1A, the plasmid DNA of pLVC18L-Npt2-SacB/R-recA and the genomic DNA of GV2260-SacB/R, but not the wild type strain GV2260, can amplify the SacB-SacR gene and a tetracycline resistance gene located on the suicide vector pLVC18L-Npt2-SacB/R-recA. All three DNAs can amplify the recA gene fragment. These results suggest that GV2260-SacB/R is carrying the SacB-SacR gene cassette. The mutant strain GV2260-SacB/R is expected to carry a non-functional recA gene (Lovett et al., 1993; Bi and Liu, 1994). In the future, it will be interesting to test the stability of plasmids maintained in GV2260-SacB/R.

To test the sucrose-sensitivity of GV2260-SacB/R, the mutant strain along with the wild type strain GV2260, and E. coli strain DH5α carrying pLVC18L-Npt2-SacB/R-recA were grown on LB agar medium supplemented with or without 5% sucrose. The wild type A. tumefaciens strain GV2260 grew equally well on LB medium with or without 5% sucrose, which suggests that sucrose in LB agar medium has no inhibitory effect on A. tumefaciens GV2260 (Figure 1B). In contrast, the E. coli strain DH5α carrying pLVC18L-Npt2-SacB/R-recA and GV2260-SacB/R grew well on LB agar medium without 5% sucrose, but showed almost no growth on LB agar medium with 5% sucrose. This result suggests that E. coli carrying pLVC18L-Npt2-SacB/R-recA and GV2260-SacB/R containing the SacB-SacR gene cassette can be effectively inhibited by the 5% sucrose presented in the culture medium.

To examine if GV2260-SacB/R can be used for plant cell transformation, the GV2260 and GV2260-SacB/R strains carrying plasmid construct pEarleygate101-YFP-HA were infiltrated into the leaves of N. benthamiana. In this construct, the YFP gene was cloned behind the CaMV 35S promoter, and it can be expressed in the transformed tobacco plant cells. Strong YFP fluorescence signals were detected from leaves inoculated with either GV2260 or GV2260-SacB/R carrying the YFP gene (Figure 2), which suggests that both strains can successfully transform the N. benthamiana plant cells.

To test the transformation efficiency of GV2260-SacB/R, N. benthamiana tissue culture and transformation was conducted with Agrobacterium strain GV2260 or GV2260-SacB/R carrying plasmid pgRNA-NbWRKY70. The construct pgRNA-NbWRKY70 carries a synthesized Cas9 gene driven by the Arabidopsis Ubiquitin 10 promoter. The expression of a guiding RNA targeting on the N. benthamiana WRKY70 gene was driven by the Arabidopsis U6 promoter. We modified the N. benthamiana leaf disk transformation protocol (An, 1985), where the leaf disks infected with Agrobacterium strains were only slightly washed, which usually can cause Agrobacterium overgrowth problems during the selection of transformed plant cells. The infected N. benthamiana leaf disks were cultured on selection medium supplemented with either 3 or 5% sucrose. The leaf-disk contamination caused by the overgrowth of Agrobacterium was recorded after 4 weeks of culture on the selection medium (Figure 3A). Under the test conditions, the contamination rates caused by GV2260-SacB/R on medium supplemented with 5 and 3% sucrose were 13.0 and 26.9% respectively, which are significant lower than the contamination rate of >80% caused by GV2260 (Figure 3B). Therefore, GV2260-SacB/R can be efficiently inhibited by the sucrose presented in the N. benthamiana tissue culture medium. However, under our testing conditions, there were still quite high numbers of contaminated leaf disks when in infections with GV2260-SacB/R. We speculate that the slightly rinsed leaf disks carried relatively high numbers of Agrobacterium cells, which may develop mutations on the SacB-SacR genes during the tissue culture process. It will be interesting to further test with different wash conditions, which may reduce the carry-on bacterium cells, and allow for further reduction of the contamination ratio. It will also be interesting to test if we can reduce or even eliminate antibiotic during the tissue culture process.

After 2 months of culture selection, putative transgenic N. benthamiana plants were generated. The number of transgenic plants generated from GV2260-SacB/R was significantly higher than those generated from GV2260 (Figure 3C) because of the lower contamination rate caused by the GV2260-SacB/R strain. Four transgenic plants generated by GV2260-SacB/R were genotyped, showing the presence of the Cas9 gene (Figure 4A). To examine if the pgRNA-NbWRKY70 transgenic plants carry mutations in WRKY70, we amplified an NbWRKY70 DNA fragment carrying the guiding RNA targeting site. All four putative transgenic lines amplified an NbWRKY70 DNA fragment with similar size (Figure 4B). The PCR products were gel purified and sequenced. As shown in the chromatogram, the PCR product amplified from the wild type plants yields a clean sequence, while the PCR product from a transgenic plant (line 3) yields double peaks near the PAM site (CGG) (Figure 4C) (Li et al., 2011; Nekrasov et al., 2013). The double peak near the PAM site indicates the heterozygosity of template DNAs, which suggests there are Cas9 induced mutations at the WRKY70 gene. The PCR products were also cloned and individual clones were sequenced, which confirmed the presence of mutants (data not shown). The phenotype of transgenic plants will be further characterized in the future. Nevertheless, our result demonstrated that GV2260-SacB/R could successfully transform N. benthamiana plant cells to generate stable transgenic plants. It will be interesting to test the transformation capacity of GV2260-SacB/R in other plant species. In this study, we also confirmed that the CRISPR-Cas9 system is a powerful tool for introducing mutations on target genes in N. benthamiana (Nekrasov et al., 2013; Belhaj et al., 2015).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.