Cell Penetrating Peptides Mediated DNA-Free Transfection in Arabidopsis and Chinese Cabbage

Background: The third-generation genome editing system CRISPR/Cas had shown strong application prospects in crop genetic improvement. However, this technology largely depends on genetic transformation. Public concerns on GMO (genetically modied organisms) safety, as well as related regulations, have restricted the application. Therefore, establishing DNA-free transfection system is important to promote CRISPR/Cas genome editing in agriculture. Results (cid:0) In this paper, cell penetrating peptides fusion protein (CPP-mCherry) was found to be effective on DNA-free transfection. DNA sequences of nine tandem arginine (R9), one cysteine (cys), reporter mCherry and histidine label were sequentially constructed into pET 45B+ expression vector and transformed into Escherichia coli BL21(DE3) strain. CPP-mCherry fusion protein can be induced by 1mM IPTG for at least 1 hour in 28 °C. CPP-mCherry fusion protein can be obtained by 200W ultrasonication, then puried by Ni column and MWCO dialysis. The Arabidopsis thaliana root tips and leaves, as well as Chinese cabbage microspores and 3-week-old microspore embryo can be used as transfected recipient. Concentration can be selected between 10-100μg/ml and incubated overnight at room temperature. R9-cys-mCherry protein can be translocated into the nucleus of microspore. The transfection eciency of root tips reached 100% and of microspore and MDE was 8.13% and 94.79%, respectively. Conclusions: Here, a CPP mediated DNA-free transfection system was built in dicots. These results lay a technical foundation of DNA-free genome editing.


Background
In recent years, the third generation of site-speci c genome editing system (CRISPR) is becoming one of the core technologies in molecular breeding. However, this technology largely depends on the genetic transformation system. Public concerns on the safety of GMOs and relevant laws restrict the application. Therefore, DNA-free transfection is of great signi cance for promoting the application.
Cell penetrating peptides (CPPs) belong to a special class of short peptides, which generally contains several amino acids. It was rst discovered in 1988 in the study of Tat protein of HIV-I virus (Frankel et al., 1988). Since then, researchers have found that some peptides also have the ability of intracellular delivery. Although these peptides are short, they contain various hydrophilic or hydrophobic motifs, which is helpful for transmembrane (Derakhshankhah et al., 2018).Cell penetrating peptide is an ideal can combine with exogenous DNA non-covalently in vitro, forming a cargo complex, and enter the recipient cells by endocytosis. The exogenous DNA is released in cytoplasm and eventually integrated into the genome Chugh et al.,2009Chugh et al., , 2010 . The CPP synthesis service is tradable. On the other hand, in order to form a covalent protein, the DNA sequence of CPPs can also be expressed together with target gene in prokaryotic system, forming a complex by disul de bridge in Vitro, and eventually transfected into the recipient cells. Bilichak carried out a bene cial attempt to transfect wheat microspore with CPP-mCherry fusion protein and eventually edit the genome by delivering ZNF protein (Bilichak et al., 2015(Bilichak et al., , 2020 Cell penetrating peptide fusion protein is an alternative method for CRISPR RNPs DNA-free intracellular delivery. Ramakrishna fused Cas9 with 9 consecutive arginine (9R) and then incubated with sgRNA to form a transfection complex. With the help of 9R, human HEK293T was successfully edited (Ramakrishna et al., 2014). In recent years, the research on DNA-free gene editing is in ascendant (Svitashev et al., 2016;Liang et al., 2017;Ran et al., 2017;Bilichak et al., 2020).
In order to develop a DNA-free transfection system on dicotyledon, Arabidopsis somatic cells and Chinese cabbage microspores and MDE were used as experimental materials in this study. We offered a DNA-free transfection method mediated by cell penetrating peptide, that can be used in DNA-free genome editing in the future.

CPP-mCherry constructs
The structures of CPP vectors used in this study are shown in Fig. 1. The R9-cys-mCherry, cys-mCherry DNA sequences are guided by T7 promoter and built in Pet45b + backbone sequence. The constructs were transformed into BL21 (DE3) competent cell.
2 Induction of CPP-mCherry fusion protein 5 µl glycerol stock were activated in 10 ml LB medium containing 0.5% glucose and 100 mg / L carboxybenzyl penicillin, pre cultured overnight at 28 °C. On the next day, 2 ml of pre culture was added to 20 ml LB medium, to make the OD value reaches 0.5, kept culture at 28 °C. In the IPTG gradient experiment, 0,0.5,1.0,1.5,2.0 mm IPTG were presented. The samples were collected after 5 h induction.
Centrifuge immediately, resuspend the bacterium in the equilibrium lysis buffer B (hispur Ni NTA spin columns kit). Then, soni cation with 200 W for 10 s on ice, repeat 6 times. After 15 000 g centrifugation for 15 min, the supernatant was the crude protein sample for SDS-PAGE detection.
3 Puri cation of CPP-mCherry fusion protein CPP-mCherry-6xHistag fusion protein was puri ed by Ni column, which was taken out from 4 °C and balanced at room temperature for 20 min before use. Ni column was centrifuged at 700 g for 2 min to remove the storage liquid. Make sure the equilibrium solution fully contacts with the column, centrifuge 700 g for 2 min to remove the equilibrium solution. The crude protein from bacterial was added into the Ni column, fully mixed, and incubated at room temperature for 30 min. 700 g centrifuge for 2 min. Rinse the Ni column with washing buffer, centrifuged at 700 g for 2 min. Elusion buffer was added to replace the target protein from the Ni column. 700 g centrifuged for 2 min. The ltrate was the target protein solution. 280 nm OD value was detected to determine the concentration by Bicinchoninic acid (BCA) method, then measured by SDS-polyacrylamide gel electrophoresis (SDS-PAGE). Non target proteins were removed by MWCO dialysis column.

Plant material
The mesophyll, root tip of Arabidopsis 'Col-0' and microspores of Brassica rapa. 'FT' were used as transfection targets. 'FT' is a double haploid (DH line) from microspore culture. Plantlets were grown in kilma, an arti cial climate zone in Wageningen University Research Center (WUR). 16  The microspore isolation protocol is modi ed from Sato et al., (1989). Buds of 'FT' were selected from healthy plantlet for surface-sterilize. After rinsing, crush the buds to press out the microspores in 5 ml NLN liquid medium. removed the debris by nylon mesh, the yellow precipitates (microspores) were used as transfection target. 7 DNA-free transfections R9-cys-mCherry fusion protein was used for DNA-free transfection, and cys-mCherry without cell penetrating peptide was used as negative control. Fusion protein were diluted as concentration gradients: 0, 10, 50, 100 µg / ml. The transfection target tissue was transferred to 1.5 ml centrifuge tube, discarded the medium, incubate with CPP-mCherry gradient solution overnight at room temperature. Centrifuge at 5000 rpm for 3 min, remove the protein solution, add 100 µl Renaissance solution to stain the cell wall for 1 h, then x the material with MTSB solution overnight at 4 °C. Transfected material was Z stacked by Leica dm5500 Q laser confocal microscope.

IPTG induction of CPP-mCherry
The induction of CPP-mCherry expression was conducted by IPTG gradient, the results are shown in Fig. 2. The obvious pink mCherry protein was clearly visible in the bacterial precipitation ( Fig. 2A). The bacterium was then broken by sonication to release the protein. The molecular weight of R9-cys-mCherry protein was about 30 kD as expected (Fig. 2B). With the increase of IPTG concentration, there was no signi cant difference in the expression of target protein (Fig. 2B).

SDS-PAGE
The target protein with 6x histag was puri ed by Ni column. After SDS-PAGE electrophoresis, the results are shown in Fig. 3. Lane 1-3 were from BL21 (DE3) competent cell as negative control. Lane 4-6 are from CPP-mCherry transformants. The negative control had no target protein as expected. In contrast, the transformants contain large amount of CPP-mCherry fusion proteins. SDS-PAGE results suggest that the cys-mCherry and R9-cys-mCherry protein could be induced from 1.0 mM IPTG induction for 5 h. After Ni column puri cation, the non-target protein was removed by MWCO dialysis.
Protein concentration was detected by Bicinchoninic acid (BCA) method. The concentrations of R9-cys-mCherry and Cys-mCherry protein were about 0.13 mg/ml (Table 1), which is su cient for the following experiments. In the beginning, 7-days Arabidopsis mesophyll cells were found to be transfected by R9-cys-mCherry accidentally (Fig. 4A). Leaves were incubated with 30 µg/ml R9-cys-mCherry overnight at room temperature. In the second day, strong mCherry red uorescence signal was observed in mesophyll cells, There is no mCherry signal in low concentration treatment. When R9-cys-mCherry concentration increased to 100 µg/ml, strong mCherry signal was found in root tip cells (Fig. 5B). No uorescence signal was found in negative control, as expected (Fig. 5A). The results suggest that the transfection shows dosedependency, the concentration of fusion protein can be selected between 10-100 µ g / ml.
In order to further determine R9-cys-mCherry distribution in recipient cell, we used high resolution confocal microscopy, Fig. 6A. In the cys-mCherry negative control, no uorescence signal was found in root tip cells Fig. 6B. In contrast, the strong red uorescence signal was found in the R9-cys-mCherry treatment. These results indicate that R9-cys-mCherry can be transfected into root tip cells in the range of 10-100 µg/ml, with 100% transfection e ciency, Table 2. DNA-free transfection was also carried out in Chinese cabbage microspores. The microspores are nearly spherical cells with thick exine, different orientation would interfere laser scanning. Therefore, the "faceup" microspores were selected for observation, whereby germination furrows are vertical to the glass slide. The thick exine was stained by Renaissance dye. The concentration gradient of R9-cys-mCherry and cys-mCherry was set as 1, 10, 50, 100 µ g / ml, respectively, Fig. 6. No cys-mCherry signal was detected at any concentration as expected (Fig. 7A). In contrast, the positive results were found in R9-cys-mCherry transfection. With the increase of concentration, strong red uorescence signals were found at 50 and 100 µg/ml (Fig. 7B). These results suggest that 'FT' microspore can be used as DNA-free transfection recipient cell. The concentration can be selected between 10-100 µg/ml. The transfection e ciency was about 8.13% (Table 3). CRISPR RNP protein complex needs to enter the nucleus, then performing gene editing. In order to clarify the distribution of R9-cys-mCherry in microspores, we observed transfected microspores by high resolution confocal, Fig. 8. Strong uorescence was clearly visible in the nucleus of microspore, indicating that CPP-mCherry was successfully transfected into microspore nucleus.
DNA-free transfection of microspore derived embryos (MDE) was carried out in this study. The concentration of CPP-mCherry was 100 µg / ml. After incubation overnight at room temperature, confocal microscopy was performed, Fig. 9.
With 100 µg / ml of R9-cys-mCherry incubation, strong red uorescence signal was found in MDE epidermal cell, as the medium section is too thick for the laser went though. MDE treated with cys-mCherry had no uorescence signal as expected. In general, CPP-mCherry can be used as a DNA-free transfection tool in crops. Microspore and 3-week-old MDEs, as well as Arabidopsis somatic cells, could be used as transfection recipient. . In this study, the latter method was utilized. The fusion protein vector mainly includes T7 promoter, cell penetrating peptide, target protein (mCherry used in this study) and puri cation tag. Pet series vectors contain T7

Discussion
promoter from λ phage, which can be used in BL21 (DE3) strains for prokaryotic expression (Yoon et al., 2009). In this study, several vectors were used for the expression of CPP-mCherry, and pet 45 + R9-cys-mCherry was the most productive one (data not shown). The location and numbers of cell penetrating peptide in the fusion protein are crucial for the transfection effect. In the vector, cell penetrating peptide can be placed at the C-terminal or N-terminal (Ru et al., 2013). The more CPP appears on the surface of the fusion protein, the more favorable it is to cross the cell membrane (Numata et al., 2012;Ng et al., 2016). CPP is located on C terminal in this study, and our transfection result is in line with previous reports.
CPPs mediated transfection is also an important method for DNA-free genome editing . We found that root tips, mesophyll cells, microspores, 3-week-old microspore embryos could be used as transfection target, which offered more types of transfection targets (Ramakrishna et al., 2014;Bilichak et al., 2015). The nuclei localization of transfection complex is important. Only when the complex was transfected into the nucleus could it perform gene editing. Arginine rich peptides (such as tat, R9) can interact with nuclear pore complexes (NPCs), and nally take the complex into the nucleus (Wender et al., 2000;Chugh et al., 2007Chugh et al., , 2010. Different from previous reports (Bilichak et al., 2015), we found that R9cys-mCherry fusion protein could be transfected into 'FT' microspore nucleus, not only on the surface (Fig. 8). Maybe because the structure and chemical composition of microspore exine are different between Cruciferae and Poaceae. There are three germinal furrows on 'FT' microspore, where the exine is thin. During microspore culture, some have double or even three furrows rupture, and most are prone to single rupture (Tang et al., 2013). R9 is a typical cationic peptide, but sporopollenin on the exine is negative charge (Salter et al., 2002;Quilichini et al., 2010;Bilichak et al., 2015). The R9-cys-mCherry would rst accumulate on the exine (Fig. 7), then enter the microspore through the furrows rupture, and nally located to the nucleus (see Fig. 8).
Cell penetrating peptide mediated DNA-free transfection has obvious advantages. Our results indicate that CPPs is an ideal DNA-free transfection tool to transfer the target protein into the nucleus, that provides theoretical and technical basis for DNA-free CRISPR RNP intracellular delivery and futher genome editing.

Conclusion
CPP-mCherry fusion protein is an e cient DNA-free transfection complex. After incubation at room temperature, the R9-cys-mCherry can be successfully transfected into somatic cells of Arabidopsis thaliana and fresh microspore, as well as MDE, of Chinese cabbage.