Arabidopsis thaliana Wassilewskija plants were used to produce calli. Root segments from Arabidopsis plants were incubated on the callus inducing medium (CIM) for four weeks to form calli. The obtained calli were then ground into small calli (approximately 1 mm in diameter) in a sterile plastic bag with the bottom of a glass bottle and suspended in 10 mL of MS medium; then, one mL of this callus solution was spread onto the CIM. The calli were subcultured every two-week. The calli were grown at 22 °C under continuous light conditions.

The Murashige-Skoog (MS) medium contained MS salts (FUJIFILM Wako Pure Chemical Corp., Japan), Gamborg’s B5 vitamins (Nacalai Tesque, Japan) 0.05% MES (FUJIFILM Wako Pure Chemical Corp., Japan), 1% sucrose, and CIM contained Gamborg’s B5 salts (FUJIFILM Wako Pure Chemical Corp., Japan), 0.05% MES, 2% glucose, Gamborg’s B5 vitamins, 2 µM 2,4-D, and 0.45% gellan gum. The shoot inducing medium (SIM) contained MS medium, 12.5 µM 2-ip, and 0.45% gellan gum.

The sequences of all primers used in this study are shown in Table 1.

The VL arm (121681-124216) located downstream of the virulence region in pTiEHA101 (Genbank accession: NZ_KY000035) was amplified by PCR from pTiEHA101 using primers, RI-Sbf-L3 F and BamHI-L3 R, then digested with EcoRI and BamHI. The EcoRI-BamHI fragment was inserted between the EcoRI and BamHI sites in p15AGm (Genbank accession: LC792546), resulting in p15AGm::VL. Then the cos site which had been amplified by PCR from bacteriophage $\lambda$ using primers, SbfI-cos F and HindIII-cos R followed by digestion with EcoRI and HindIII was inserted between HindIII and EcoRI sites in p15AGm-VL, resulted in p15AGm::VL/cos. The VR arm (85498-87041) located upstream of the virulence region in pTiEHA101was amplified from pTiEHA101 using primers, R3-Bg F and R3-Pst R and cloned into pMD20 T-vector (Takara Bio Inc.), giving pMD::VR. In order to insert the cos site to pMD::VR, the cos sequence was amplified using primers, SbfI-cos F and HindIII-cos R followed by digestion with SbfI and HindIII was cloned between SbfI and HindIII in pUC19. The VR region excised with EcoRI and XbaI from pMD::VR was inserted between EcoRI and XbaI sites in pUC19 with the cos sequence, resulted in pMD::VR/cos. The repABC sequence with the mutation Y311H in repB (repH) was generated by fusion PCR. The 5^′ part of repH was amplified by PCR using primers BamHI-repABC F and Y311H R from pTiEHA101 and the 3^′ part was using primers, Y311H F and SpeI-repABC R. At the second PCR, the 5^′ part and the 3^′ part were mixed and subjected as templates the fused fragment (repH) was amplified using primers BamHI-repABC F and SpeI-repABC R. The repH fragment was digested with BamHI and SpeI and inserted between BglII and XbaI in pMD::VR/cos, resulted in pMD::VR/repH/cos. Wild type repABC gene was also amplified from pTiEHA101 using the primers, BamHI-repABC F and SpeI-repABC R. After digestion with BamHI and SpeI, the wild type repABC was also inserted between BglII and XbaI in pMD::VR/cos to generate pMD::VR/repABC/cos. To remove the VR region from pMD::VR/repABC/cos and pMD::VR/repH/cos, the BamHI-AgeI region including the VR arm and the N-terminal part of repABC or repABC in pMD::VR/repABC/cos or pMD::VR/repH/cos were replaced with the PCR fragments amplified from pMD::VR/repABC/cos, BamHI-repABC F and SpeI-repABC R followed by BamHI and AgeI digestion, resulting in pMD::repABC or pMD::repH, respectively. p15AGm::VL and pMD::VR/cos were sequentially transformed into EHA101 and integrated into pTiEHA101 by homologous recombination, resulting in pEHA101/VLcos/VRcos. pEHA101/VLcos/VRcos was purified from the Agrobacterium cells and was subjected to $\lambda$ packaging using Lambda-inn in vitro packaging kit (Code No. 317-0741, Nippon Gene Co., Ltd., Japan), then the reacted solution was infected to E. coli strain DH5$\alpha$. The resulting plasmid, pTiBoVIR was purified from the E. coli cells and transformed into C58C1 carrying pMD::VR/repH/cos and they were integrated by homologous recombination, resulting in pInt-repH/vir/cos. pInt-repH/vir/cos was digested with SbfI, then self-ligated and transformed into DH5$\alpha$, giving pCU307D. The expression cassette of NLS-GUS was excised with HindIII and EcoRI from pBI-NLS/GUS (Genbank accession: PV369457) and inserted between HindIII and EcoRI in pTK222 (Genbank accession:PV754030), resulting in pTES::NLS-GUS. In order to construct pRK2-A281virF, the A281virF gene was amplified by PCR using primers, XhoI-A281virF F and AvrII-A281virF R from pTiEHA101 and then digested with XhoI and AvrII. The vector backbone was amplified by PCR using primers, XbaI-trfA F and PlacIq R from pT5exRKH (Genbank accession: LC792545), then digested with XbaI and XhoI. pRK2-A281virF was generated by ligating these fragments.

Twenty ng of total DNA extracted from EHA101 (C58C1 carrying pTiEHA101), C58C1 carrying pMD::repABC, C58C1 carrying pMD::Y311H and C58C1 carrying pCU307D was used as a template for the PCR. PCR was performed on the CFX96™ Real-Time PCR Detection System (BIO-RAD, USA) with Premix Ex Taq™ (Probe qPCR) (RR390A, Takara, JAPAN) according to the manufacturer’s instruction. The gene-specific primers were designed to produce 126 (primers; virG qF and virG qR), 117 (primers; repB CF and repB CR) and 146 (primers: Atu0972F and Atu0972R) bp DNA fragments from the genes of virG, repB in Ti-plasmids and Agrobacterium Atu0972 gene used as an internal control, respectively. The sequences of the primers are shown in Table 1. Relative copy numbers were calculated by normalization with those of the Atu0972 gene.

Arabidopsis calli were freshly broken into fine pieces as described in 2.1. Agrobacterium cells were co-cultured with the crushed calli for 48 hours, then the calli were washed with sterilized water three times. The washed calli were incubated on CIM with 100 µg/mL of carbenicillin for a further 24 hours to allow transient expression in plant cells.

The expression assay of GUS reporter gene was done by GUS histochemical assay and MUG assay in Arabidopsis cells. For GUS histochemical assay, the transiently transformed Arabidopsis cells were incubated at 37 °C for 4 hours in the GUS staining solution (50 mM Na₃PO₄ pH 7.2, 0.5 mM K₃[Fe(CN)₆], 0.5 mM K₄[Fe(CN)₆], 0.1% Triton X-100, 0.5 mg/mL 5-bromo-4-chloro-3-indolyl glucuronide (X-gluc), 2mM MgCl₂). After incubation, the buffer was removed and added with a stop buffer containing 50 mM MES and 50 mM KCl. Microscopic analysis was done to determine differences between Agrobacterium strains. For the MUG assay, the Arabidopsis calli were added with 1x Tris Buffered Saline (TBS) and 0.1% Triton x-100. These were squeezed using the disposable homogenizer. This was centrifuge at 15,000 rpm for 10 minutes and the supernatant was transferred into a new tube. After, 5 µL of the extract was incubated at 37 °C for 1 hour in 50 µL 4-methylumbelliferyl $\beta$-D-Glucuronide (MUG) assay buffer (50 mM Sodium phosphate pH 7.0, 1 mM MUG, 10 mM EDTA, 10 mM $\beta$-mercaptoethanol, 0.1% Triton x-100). The reaction was stopped by adding 945 µL of 0.2 M Na₂CO₃. Fluorescence was determined using 25 µL of the solution in microplate reader with excitation at 365 nm and emission at 455 nm.

We used the Arabidopsis ecotype Wassilewskija (WS) for our tissue culture experiments, as Czakó et al. [10] have reported that cultured WS cells have the highest efficiency of regenerating shoots among the examined ecotypes. It is known that long-term incubation of Arabidopsis calli on CIM containing 2,4-D inhibits shoot regeneration in SIM containing cytokinin [11]. Our previous study demonstrated that a high cell density of cultured cells caused by extended incubation on CIM had an inhibitory effect on shoot regeneration [12]. To confirm the shoot-regenerating ability of calli that had been maintained for a long time, calli derived from Arabidopsis WS root segments by culturing on CIM were subcultured every two weeks for 26 months. Ground calli (see 2.1) were then grown on CIM for two weeks and then transferred onto SIM (Fig. 1). A large number of shoots were formed on calli that had grown larger. More shoots formed on parts of the calli in contact with the medium (Fig. 1C) compared to those on aerial parts. These results clearly showed that calli that had been subcultured for a long period retained their ability to regenerate shoots on SIM. We chose Arabidopsis calli as plant materials for transient expression experiments since they can be able to be used for model experiments for genome editing.

Fig. 1.

Shoot regeneration from Arabidopsis calli. Routinely subcultured calli on day 14 after the last subculture were transferred to shoot inducing medium (SIM). Images show identical calli on the SIM. (A) Calli on day 0 after transfer to SIM. (B,C) Calli on day 14 after transfer to SIM. Plate with the calli on the indicated day after transfer to the SIM. Each picture was taken from the table side (B) or the other side (C). (D) Area with black frame in (B). Bar = 2 mm.

In order to improve the efficiencies of T-DNA transfer by Agrobacterium, we attempted to increase the copy numbers of Ti-plasmids in Agrobacterium cells. Ti-plasmids are megaplasmids with sizes of 200–250 kbp and have low copy numbers. The replication origins of Ti-plasmids are the repABC operons which are commonly found in alfaproteobacteria plasmids [13, 14]. EHA101 is called the hypervirulence strain [15] and we chose this strain as the initial material to manipulate. It had been reported that a mutation, Y299H in the repB gene in pRiA4b increased the copy numbers of the plasmids and the alignment of amino acid sequences from several repABC revealed that the tyrosine residue was conserved among a variety of repB proteins [16]. We introduced the same amino acid change, tyrosine to histidine in the repB gene in pTiEHA101 (originated from pTiBo542) (Fig. 2A). Fig. 2B,C shows the comparison of copy numbers in pTiEHA101/C58C1 (common strain name, EHA101), pMD::repABC/C58C1 (wild type repABC gene) and pMD::repH (repH: repABC with the mutation Y311H), determined by qPCR. The copy numbers of pMD::repABC were approximately 7.2-fold higher than those of pTiEHA101, probably owing to the difference of their sizes. The size of pMD::repABC is 6.8 kbp, while that of pTiEHA101 is 189 kbp, suggesting plasmids with smaller sizes give more copy numbers compared to those with larger sizes and the identical replication origin. In addition, introduction of the mutation, Y311H gave approximately twice copy numbers, compared with wild-type repABC and 14.2-fold, and 7.2-fold, compared with pTiEHA101 and pMD::repABC, respectively. The increased ratios caused by the Y311H mutation were much lower (approximately 2.0-fold) than the Y299H mutation (approximately 10-fold) in pRi repABC [16] when plasmids with the same sizes were examined. The effects on copy numbers by reduced sizes were much greater than those by the Y311H mutation.

Fig. 2.

Relative copy numbers of pTiEHA101-derivatives in Agrobacterium cells. (A) Alignment of wildtype RepB protein sequences in pRiA4b and pTiEHA101 corresponding to mutation sites Y299H and Y311H. (B) Schematic structures of plasmids used in the experiments. EHA101; pTiEHA101, pMD::repABC; a small-sized plasmid with wild type repABC, pMD::repH; a small-sized plasmid with repH. (C) The relative copy numbers were determined by qPCR. The vertical axis shows the plasmid copy number. Relative copy numbers are shown as normalized values to those of EHA101. Experiments were individually performed three times. Error bars represent standard error of the mean.

These results led us to make the size of the helper Ti-plasmid reduce by cloning only the virulence (vir) region. According to information reported previously [15], we designed the strategy to clone the vir region by introducing two cos sites at the ends of the vir region, a selectable marker gene and the p15A replication origin for E. coli, then packaging as $\lambda$ phage particles since its size was close to $\lambda$ phage genome. As shown Fig. 3, pTiBoVIR (approximately 41 kbp) was successfully constructed. Then the repH operon was also integrated into pTiBoVIR, resulting in pCU307D. The copy numbers of pCU307D in Agrobacterium strain, C58C1 were 2.8-fold higher compared with those of pTiEHA101 in C58C1 (EHA101) (Fig. 4). Their T-DNA transfer efficiencies were examined by transient expression of $\beta$-glucuronidase (GUS) in Arabidopsis calli after co-culture with EHA101 or CU307D (C58C1 with pCU307D) carrying the reporter plasmid, pTES::NLS-GUS. The GUS coding sequence with a nuclear localization sequence at its N-terminal to prevent the products from diffusion to adjacent cells through plasmodesmata was used for transient expression in plant cells (Fig. 5A). GUS activities in Arabidopsis cells after co-culture with EHA101 or CU307D carrying pTES::NLS-GUS were similar against our expectation that higher copy numbers of the vir genes would confer more efficient T-DNA transfer (Fig. 5B), although these results demonstrated that indeed, only the vir region was sufficient to transfer T-DNA into plant cells.

Fig. 3.

Schematic diagram of pCU307D construction. Abbreviations indicate the followings: cos, the recognition sequence by lambda terminase; AmpR, the ampicillin/carbenicillin resistance gene; VR, homologous sequence to pTiEHA101 located at 5^′ of the vir region; VL, homologous sequence to pTiEHA101 located at 3^′ of the vir region; RepH, the replication origin for Agrobacterium cell; SbfI, the recognition sequence by SbfI; p15 ori and pUC ori, the replication origins for E. coli cells.

Fig. 4.

Determination of relative copy numbers of pCU307D. The relative copy numbers were determined by qPCR. Relative copy numbers were calculated as normalized values to those of EHA101. Experiments were individually performed three times. Error bars represent standard error of the mean.

Fig. 5.

T-DNA transfer efficiencies by EHA101 and CU307D. (A) A schematic diagram of the GUS expression construct where the T-DNA region is flanked by Right Border (RB) and Left Border (LB). The T-DNA transfer efficiencies were examined by transient expression of $\beta$ -glucuronidase (GUS) in Arabidopsis cells. The construct contains the cauliflower mosaic virus 35S promoter (P35S) driving expression of the $\beta$-glucuronidase (GUS) coding sequence fused to a nuclear localization signal (NLS-GUS) followed by the nopaline synthase terminator (Tnos). (B) GUS activity in Arabidopsis cells co-cultured with EHA101 and CU307D carrying pTES::NLS-GUS was quantified using the MUG assay. Experiments were individually performed three times. Error bars represent standard error of the mean.

F-box protein encoded by the virF gene in pTiC58 has been reported to enhance transformation efficiencies by Agrobacterium strain, C58 [17]. The virF gene is located in the vir region of pTiC58, while a similar gene encoding an F-box protein (165435-166301 in Genbank accession: NZ_KY000035) in agropine-type Ti-plasmids is located outside of the vir region [9]. In order to examine the effects of the A281virF gene encoding the F-box protein in pTiEHA101 (derived from pTiBo542 in Agrobacterium strain, A281) on T-DNA transfer, a plasmid carrying the gene, pRK2-A281virF was introduced into C58C1 with pCU307D, since it had been eliminated during construction of pCU307D from pTiEHA101. Introduction of pRK2-A281virF into C58C1 carrying pCU307D clearly enhanced transient expression of GUS in plant cells to 3.2 higher (Fig. 6) and the rates are similar to those of their copy numbers. We were not able to judge that the enhancement was caused by more efficient infection by Agrobacterium cells or larger amounts of T-strand transportation from the pictures of GUS staining. We named C58C1 carrying pCU307D and pRK2-A281virF as CU307DF.

Fig. 6.

Effects of the A281virF gene on T-DNA transfer efficiencies. Transient expression of GUS was examined after co-culture of Arabidopsis cells and Agrobacterium strains indicated. All strains carry pTES::NLS-GUS shown Fig. 5A. (A) The A281virF gene was cloned into a vector with RK2 (the replication origin for Agrobacterium) and HPT (the hygromycin resistance gene). (B) Transient expressions of GUS in Arabidopsis cells were visualized by X-gluc staining. Scale bars indicates 100 µm. (C) Quantitative GUS activities in Arabidopsis calli. Experiments were individually performed three times. Error bars represent standard error of the mean.