A method using electroporation for the protein delivery of Cre recombinase into cultured Arabidopsis cells with an intact cell wall

Genome engineering in plants is highly dependent on the availability of effective molecular techniques. Despite vast quantities of research, genome engineering in plants is still limited in terms of gene delivery, which requires the use of infectious bacteria or harsh conditions owing to the difficulty delivering biomaterial into plant cells through the cell wall. Here, we describe a method that uses electroporation-mediated protein delivery into cultured Arabidopsis thaliana cells possessing an intact cell wall, and demonstrate Cre-mediated site-specific recombination. By optimizing conditions for the electric pulse, protein concentration, and electroporation buffer, we were able to achieve efficient and less-toxic protein delivery into Arabidopsis thaliana cells with 83% efficiency despite the cell wall. To the best of our knowledge, this is the first report demonstrating the electroporation-mediated protein delivery of Cre recombinase to achieve nucleic acid-free genome engineering in plant cells possessing an intact cell wall.

In order to introduce genetic materials into plant cells through the cell wall, the delivery of DNA using a particle gun or Agrobacterium (Rhizobium)-mediated techniques are widely used in plant research 25,26 . Although these methods enable the efficient expression of genome engineering proteins, including Cre recombinase, exogenous DNA fragments that are introduced using a particle gun or Agrobacterium are often or invariably incorporated into the genomic DNA, and may induce unexpected effects that interfere with subsequent analyses. In addition, the success of Agrobacterium-mediated delivery is strongly dependent on the combination of the bacterial strain and the plant species or variety. To overcome these obstacles, a method for the delivery of proteins directly into plant cells possessing a cell wall is in high demand. Several groups have attempted the delivery of proteins into Chlamydomonas and tobacco cells through the cell wall using electroporation [27][28][29] . However, the delivery efficiency was either not quantified or was very low, and little is known about whether the delivered proteins are functionally active in the cell nucleus and cytoplasm. Although Cao et al. demonstrated the delivery of a modified Cre protein into rice calli after protoplasting or plasmolyzing and achieved regeneration of rice plants with Cre-excised target sequence 15 , the delivery efficiency was insufficient or harsh for the biological analysis of cultured plant cells.
Here, we demonstrate an efficient electroporation technique for the delivery of proteins into cultured plant cells that possess a cell wall. Using Arabidopsis thaliana as a model, we constructed a reporter cell-line that responds to Cre recombinase and then expresses the gene for ß-glucuronidase (GUS), which enables us to quantify the delivery efficiency. By optimizing the conditions for the electric pulse, protein concentration, and electroporation buffer, we successfully achieved efficient and less-toxic protein delivery in 83% of Arabidopsis thaliana cells, despite the cell wall. To the best of our knowledge, this is the first report to demonstrate the electroporation-mediated protein delivery of Cre recombinase to achieve nucleic acid-free genome engineering in plant cells possessing a cell wall.
Plasmid construction. To construct pCAMBIA-N-xGxGUS, the NOS promoter was amplified with primers (ACGGCCAGTGCCAAGCTTGATCATGAGCGGAGAATTAAG and TCTGCGAAAGCT CGACCTAGGAAACGATCCAGATCCGGTGCA) from pRI201 (TaKaRa Bio. Inc., Shiga, Japan). The resulting fragment was cloned using the In-Fusion HD Cloning Kit (Takara Bio) into pCAMBIA 1305.2 (Marker Gene Technologies Inc., Eugene, OR, United States), which had been partially digested with HindIII and XhoI. The GFP fragment (mEmerald) was sandwiched between two loxP sites, and was subsequently amplified using p ri me rs ( G  G AC TC TT GA CC AT GT AA TA AC TT CG TA TA GC AT AC AT TA TA CG AA GT TA TG TTAACTACATCACAATCA  CACAAAAC and T TA GTAGTAGCCATGGTCTAGATAACTTCGTATAATGTATGCTATACGAAGTTATGGG  CCCCTTATCTTTAATCATATTCCA) from pcDNAFRTxE2CxmEm 30 . The resulting fragment was cloned into the NcoI site using the In-Fusion HD Cloning Kit. To construct pCAMBIA-B-HNCre(A207T), the Bsd R gene was amplified using primers (ATCTATCTCTCTCGACCTAGGTTGATAGATATGGGCCAGGCCAAGCCTTTGT and TATGGAGAAACTCGATTTAAATTAGCCCTCCCACACATAACCAGA) that were originally from pTracer-EF/Bsd (Thermo Fisher Scientific, Waltham, MA, United States). The resulting fragment was cloned by In-Fusion into pCAMBIA 1305.2, which had been digested with XhoI. The Cre fragment was then amplified using primers (GGACTCTTGACCATGGGCCACCATCACCAC and ATTCGAGCTGGTCAC CCGTCGACGTTAATCGCCATCTTCCAGCAG) from pFT-HNCre(A207T), and the resulting fragment was cloned by In-Fusion between the BstEII and NcoI sites. Please also refer to Supplementary Figure 1.
Cell materials and culture. The Arabidopsis thaliana T87 cell line was obtained from RIKEN Bio Resource Center (Ibaraki, Japan) and cultured in a liquid NT1 culture medium (30 g/L sucrose, 0.1 mM KH 2 PO 4 , 1 × Murashige Skoog Salt Mixture and Vitamins, 2 µM 2,4-dichlorophenoxyacetic acid, pH 5.8 adjusted with KOH) at 22 °C while shaking under light, unless otherwise specified. Cells were maintained by 15-fold weekly dilutions. Agrobacterium tumefaciens (Rhizobium radiobacter) strain GV3101 was cultured in LB medium (Merck, Darmstadt, Germany) supplemented with 50 µg/mL rifampicin and 30 µg/mL gentamicin at 28 °C with shaking. To prepare the transformants, GV3101 cells were electroporated with the indicated binary plasmids and cultured in LB medium supplemented with 50 µg/mL rifampicin, 30 µg/mL gentamicin, and 50 µg/mL kanamycin. Prior to infection, GV3101 cells carrying the indicated binary plasmids were inoculated into LB medium supplemented with 50 µg/mL rifampicin, 30 µg/mL gentamicin, and 50 µg/mL kanamycin. After an overnight culture, cells were washed thrice with liquid B5 medium (30 g/L sucrose, 0.5 g/L MES, 1x Gamborg's B5 Salt Mixture and Vitamins, and 1 µM 1-naphthaleneacetic acid) and resuspended to a final OD 600 of 0.6.

Agrobacterium-mediated transformation of T87 cells. Cells were transferred into liquid B5 medium
and cultured for 2 days. The cells containing 30 µL of packed cell volume (PCV) were then infected with 0.8 µL of GV3101 inoculate (OD 600 of 0.6) carrying the indicated binary plasmids with 200 µM acetosyringone. Two days after infection, the cells were washed thrice using liquid B5 medium supplemented with 200 µg/mL cefotaxime to eliminate GV3101, and further cultured in liquid B5 medium supplemented with 200 µg/mL cefotaxime for 3 days. For the construction of a reporter T87 cell line, the cells were additionally transferred to CIM agar (0.6% agar, 30 g/L sucrose, 0.5 g/L MES, 1x Gamborg's B5 Salt Mixture and Vitamins, 1 µM 1-naphthaleneacetic acid, pH 5.7 adjusted with KOH) with 200 µg/mL cefotaxime and 20 µg/mL hygromycin and cultured for 2 weeks. A single clone, exhibiting bright green fluorescence, was picked and cultured in liquid NT1 medium containing 10 µg/mL hygromycin, and designated as the T87-xGxGUS cell line. T87-xGxGUS cells were maintained in the same way as T87 cells, using the liquid NT1 medium containing 10 µg/mL hygromycin.
Electroporation. T87-xGxGUS cells were 15-fold diluted into fresh NT1 medium 1-5 days before electro- Evans blue staining. Cytotoxicity was analyzed by staining with Evans blue 1 h after electroporation. Cells were washed once with water and incubated with 0.05% (w/v) Evans Blue (FUJIFILM Wako Pure Chemical, Osaka, Japan) staining solution for 15 min at RT. Cells were then washed once with water. Images were obtained using a phase contrast microscope (Nikon, Tokyo, Japan). For quantification, the supernatant was discarded, and Evans blue stain was extracted using 50% methanol 1% SDS. Absorbance at 595 nm was measured spectrophotometrically using a NanoDrop ONE (Thermo Fisher Scientific).

GUS staining.
To visualize GUS expression, cells were incubated with 0.5 mg/mL of X-Glucuronide (X-Gluc, Carbosynth, Berkshire, United Kingdom) dissolved in staining buffer (20% (w/v) methanol, 50 mM NaH 2 PO 4 , pH 7.0) for 30 min at 37 °C. For the staining of protoplasts, mannitol (final concentration of 400 mM) was added to the GUS staining solution. Cells were then transferred into 70% (w/v) ethanol and images were obtained using a phase contrast microscope (Nikon).

Fluorescence quantification of GUS activity. Two days after electroporation, cells were washed once
with NT1 medium (pH 7.0; adjusted with KOH). Cells with 5 µL of PCV were then suspended in 100 µL of NT1 medium (pH 7.0) containing 10 µM 6-chloro-4-methylumbelliferyl β-D-glucuronide (CMUG; Glycosynth, Cheshire, United Kingdom) and transferred to a black 96-well plate (Thermo Fisher Scientific). Fluorescence was measured using a Wallac 1420 ARVOsx microplate reader (PerkinElmer, Waltham, MA, United States). The excitation wavelength was set at 355 nm and the emission wavelength at 460 nm. Chlorophyll was quantified to normalize GUS activity. For the chlorophyll extraction, cells with 5 µL of PCV were immersed in N,N-dimethylformamide for over 6 h at 4 °C in the dark. Absorbance at 646 and 664 nm was measured using a NanoDrop ONE. The amount of chlorophyll was then calculated according to a previous study 31 . GUS activity is obtained as the increase in fluorescence intensity per minute per microgram of chlorophyll a and b. Quantification and Statistical Analyses. All measurements are presented as mean ± standard error (SE).
Sample sizes are indicated in the figure legends.

Results
Establishment of reporter T87 cell line. Protein delivery into cells is a promising approach for biotechnology, and thus has been extensively developed in mammalian cells. However, this technique has not been widely used in plants thus far, as the penetration of the cell wall and validating the successful introduction of protein into cells has proven to be problematic. Because of the chloroplast-a common feature of the vast majority of plants-fluorescently-labeled proteins are difficult to observe owing to autofluorescence from chlorophyll. Furthermore, the visualization of fluorescently-labelled biomaterials in fixed tissues or cells results in a substantial degree of artifacts owing to fixation 32 . Recently, Cedeño et al. electroporated a stress-related protein ERD14 into the cell cytoplasm of tobacco BY-2 cells 29 , although the microscopic observation of a fluorescent probe does not unequivocally prove that intracellular delivery was successful, as the proteins fused with fluorescent proteins do not often reflect their original localization 33 . In addition, if the protein of interest can exert its function outside of the cell nucleus or cytoplasm, such as the endosome or intercellular gap, the protein may appear but not actually localize to the cell nucleus or cytoplasm 34 . Thus, one must pay special attention to proving the intracellular delivery of biomaterials.
To verify the intracellular delivery of proteins in Arabidopsis thaliana cells, we adopted a Cre protein system, which is only active when introduced into the cell nucleus. To evaluate the activity of the intracellular delivery of Cre proteins, we first established a reporter cell line (T87-xGxGUS) by stably integrating part of pCAMBIA-N-xGxGUS-a binary plasmid encoding green fluorescent protein (GFP), followed by a transcription termination signal and a sequence encoding GUS-into the genome of the Arabidopsis thaliana T87 cell line using Agrobacterium (Fig. 1a). Once Cre recombinase is introduced into the cell nucleus and the intervening sequence between two directly oriented loxP sites is excised, T87-xGxGUS cells-originally expressing GFP-come to express GUS, which can be stained with blue dye using X-Gluc or produces a fluorescent product from CMUG (Fig. 1a). To test whether the reporter cells work, cells were infected with Agrobacterium carrying pCAMBIA-B-HNCre(A207T), a binary plasmid encoding Cre recombinase under the 35 S promoter. Three days after infection, some of the cells exhibited GUS expression upon Cre recombination (Fig. 1b). This result demonstrates that T87-xGxGUS cells enable us to validate the intracellular delivery of the Cre protein.

Electroporation of Cre protein into T87 cells with an intact cell wall. Next, we attempted to deliver
Cre protein into T87 cells with an intact cell wall. NEPA21 TypeII was used for the electroporation. NEPA21 TypeII generates a square wave pulse, which yields higher viabilities of electroporated cells in comparison to an exponential decay wave pulse. GUS activity was assessed 2 days after electroporation. Additionally, cell viability was tested by Evans blue staining 1 hour after electroporation. We tested several buffers (e.g., NT1 medium (modified MS), B5 medium, PBS, and Opti-MEM I) for electroporation. The electric resistivity of the four electroporation buffers-NT1 medium, B5 medium, PBS, and Opti-MEM I-were 1.86, 2.92, 0.850, and 0.903 Ω⋅m, respectively. As shown in Fig. 2a, there were few GUS-expressing cells when electroporation was performed using NT1 medium or B5 medium. Although some cells expressed GUS when electroporated using PBS, high cytotoxicity was observed in this condition (Fig. 2b). Surprisingly, when Opti-MEM I was used as an electroporation buffer, most cells expressed GUS with less cytotoxicity (Fig. 2a and b). As demonstrated by these experiments, we successfully accomplished electroporation-mediated protein delivery into T87 cells with an intact cell wall using Opti-MEM I. To investigate which components of Opti-MEM I promote the electroporation efficiency, we tested  Table). The presence of these components could reduce cytotoxicity. In addition, we examined the effectiveness of another electroporation system called Nucleofector. As shown in Supplementary Figure 4, Cre proteins were successfully delivered using Nucleofector when Opti-MEM I was used as an electroporation buffer.

Effect of poring pulse conditions on electroporation efficiency. As the electroporation efficiency
is largely affected by the electric condition of the poring pulse, we decided to assess the effect of poring pulse conditions using the NEPA21 TypeII electroporation system. The three main parameters of poring pulse are: field strength, duration, and the number of pulses. The standard poring pulse conditions, used in Fig. 2a, was 5 poring pulses of 375 V/cm for 10 ms. To evaluate the effect of variations in these parameters, we set the various conditions as follows: field strength (250, 375, 500, or 625 V/cm), pulse duration (2, 5, 10, or 25 ms), and the number of poring pulses (1, 2, 5, or 9). To quantify the electroporation efficiency, we measured the enzymatic activity of GUS using CMUG, and normalized it to the amount of chlorophyll. As shown in Fig. 3a, the normalized GUS activity of T87-xGxGUS cells increased concomitantly with the field strength of the poring pulse,  although cytotoxicity also increased in response to increases in field strength (Supplementary Figure 5a). While the normalized GUS activity increased concomitantly with the duration of poring pulse (Fig. 3b), cytotoxicity was nearly unchanged within the range of 2-10 ms (Supplementary Figure 5b). The number of poring pulses did not affect normalized GUS activity at more than 2 pulses (Fig. 3c), while cytotoxicity did not increase at greater than 1 pulse (Supplementary Figure 5c). These results indicate that the field strength and duration of poring pulse are important factors for protein electroporation into Arabidopsis thaliana cells, and that more protein can be electroporated by increasing the field strength or/and duration of poring pulse.
Effect of the size of cell aggregates on electroporation efficiency. In general, bacterial or mammalian cells are well-suspended by pipetting or by trypsinization to disperse the cells prior to electroporation, as surface area increases when cells are well-dispersed. Since cultured Arabidopsis thaliana cells form large aggregates as shown in Fig. 2a, aggregate size may affect delivery efficiency. To test this hypothesis, T87-xGxGUS cell aggregates were fractionated into small (<100 μm), medium (100-300 μm), and large (>300 μm) aggregates by passing through mesh of varying sizes prior to electroporation (Fig. 4a). Cells were then electroporated with 1 μM of Cre protein using the standard electroporation program and normalized GUS activity was measured. Surprisingly, the larger cell aggregates exhibited higher GUS activity, where the GUS activity of large aggregates was three times higher than that of small aggregates (Fig. 4b). The large aggregates are more likely to survive electroporation in comparison to small aggregates (Fig. 4c), consequently causing the ratio of GUS-expressing cells to increase in the large cell aggregates. This result suggests that Arabidopsis thaliana cells forming large aggregates, resembling callus or tissues, are suitable for protein electroporation.
Effect of protein concentration on electroporation efficiency. The efficiency of the intracellular delivery of exogenous materials such as DNA, RNA, protein, and low-molecular compounds is highly dependent on the concentration of these materials. Therefore, we assessed the effect of Cre protein concentration on electroporation efficiency. Cells were electroporated with 0.01-5 μM of Cre protein using the standard electroporation program. GUS staining showed that the number of stained cells increased concomitantly with the concentration of Cre protein (Fig. 5a). The quantification of fluorescence also showed that GUS activity increased with increasing concentrations of Cre protein (Fig. 5b), while cytotoxicity was slightly increased concomitantly with the concentration of Cre protein (Fig. 5c). When electroporated with more than 0.2 μM of Cre protein, the cellular expression of GUS was substantial. These results reveal that protein concentration was an important factor in the efficacy of protein electroporation into Arabidopsis thaliana cells, where more than 0.2 μM of protein was required for efficient electroporation. Furthermore, a detailed quantification of delivery efficiency was performed to clarify the population of cells in which protein had been successfully delivered. As almost all T87 cells form multicellular aggregates, cell dissociation is required to count the number of cells. To this end, we electroporated 5 μM of Cre protein to T87-xGxGUS cells (with intact cell walls), and then generated protoplasts using cellulase and pectinase to dissociate the cell aggregates. GUS staining revealed that 83% of cells expressed GUS ( Fig. 6a and b). Next, we performed a genomic PCR analysis using primers that flanked the two loxP sites in the reporter cassette (Fig. 6c). As shown in Fig. 6d, electroporated cells exhibited a 79.4% recombination rate. This result is consistent with the fact that the frequency of GUS-expressing cells in all electroporated cells is approximately 83% (Fig. 6b). These results show that we successfully accomplished the efficient and less-toxic delivery of protein into Arabidopsis thaliana cells through the cell wall using electroporation, and most notably that the delivery efficiency was as high as 83%.

Discussion
Effective techniques for the delivery of proteins of interest into cells are essential to biological research and bioengineering. The conventional method of introducing the DNA sequence for a given protein into a cell as a plasmid, and subsequently expressing the protein, has been widely accepted in the field of biology. However, vector-mediated protein expression often requires time-consuming steps, such as the optimization of codon and promoter sequences according to target cells. Furthermore, the delivered DNA frequently integrates into the host genome and can cause a variety of unintended side effects, such as unexpected gene disruption. As a result of the low level of stable integration by Agrobacterium-mediated gene delivery, transformants need to be selected using antibiotics. To address these problems, direct protein delivery-in which purified proteins are delivered into target cells without any DNA or RNA-has been attracting more and more attention. As proteins should be highly purified, this method substantially diminishes the risk of inserting exogenous DNA into the host genome, and eliminates the effort of codon and promoter optimization. Furthermore, in the case of genome engineering enzymes, as directly introduced proteins immediately modulate the target genome allow for temporary action in cells, the off-target effects of these enzymes can be reduced [35][36][37] . We have focused on methods of protein delivery and has introduced genome engineering enzymes, such as zinc-finger nuclease and Cre recombinase, to mammalian cells using direct delivery and nanoneedles 30,37,38 . Most notably, while genome engineering proteins are delivered for a relatively short period of time, the effects last permanently as the cell proliferates. Other research groups have also reported protein delivery methods for Cre recombinase, transcription activator-like effector nuclease (TALEN), and CRISPR-associated protein 9 (Cas9) 39,40 . However, to date, most of these techniques have been limited to animal cells. It has been previously thought that the direct delivery of proteins was difficult in plant cells owing to the thick cell wall surrounding the cell membrane. Although a few studies have reported the successful delivery of proteins into plant protoplasts whose cell walls were enzymatically removed by cellulase and/or pectinase, these techniques require time-consuming steps and high-cost reagents owing to the number of enzymatic reactions. Moreover, whole plant regeneration from protoplasts is difficult for many plant organisms. Therefore, we decided to establish a method for the effective delivery of proteins into plant cells with the cell wall intact. Moreover, while it is well known that cell walls also exist in fungi and bacteria such as E. coli-which has been extensively used for plasmid cloning in the field of molecular biology-the constitutions of the cell walls of these organisms differ from those of plants. In the present study, we used Cre protein as the first demonstration of nuclear delivery in plants, where the biophysical nature of the protein-such as the size (40 kDa) and surface charges (isoelectric point of 9.8)-may affect the delivery efficiency. Surprisingly, we found that Cre protein can be simply and efficiently delivered into Arabidopsis thaliana cells (with up to 83% efficiency), even in the presence of a cell wall, by immersing Arabidopsis thaliana cells in an optimized buffer and performing electroporation. Since Cre forms a tetrameric complex with DNA in the recombination reaction, at least 4 molecules should be delivered into given a cell to facilitate this recombination reaction. Although further studies are needed to clarify the mechanism by which cells take in protein by analyzing the state of the cell wall and cell membrane immediately after electroporation, our finding that proteins can be delivered into Arabidopsis thaliana cells without removing the cell wall could considerably advance the field of plant genetics.
In conclusion, Cre recombinase, which is one of the most widely used genome engineering enzymes, was delivered into Arabidopsis thaliana cells through the cell wall with a high degree of efficiency. As Cre recombinase can induce site-specific recombination, this technology will enable the conditional knockout of lethal genes and the removal of unnecessary selection markers that adversely affect cell functions in cultured Arabidopsis thaliana cells. Research using cultured cells enables us to analyze gene functions more quickly and robustly with lesser variation at the level of the individual experiments. This simple, economical, and effective technology will contribute to the biological analysis of genes and cellular functions in cultured Arabidopsis thaliana cells, which has proven difficult to analyze thus far.