Highly efficient mesophyll protoplast isolation and PEG-mediated transient gene expression for rapid and large-scale gene characterization in cassava (Manihot esculenta Crantz)

Background Cassava (Manihot esculenta Crantz) is a major crop extensively cultivated in the tropics as both an important source of calories and a promising source for biofuel production. Although stable gene expression have been used for transgenic breeding and gene function study, a quick, easy and large-scale transformation platform has been in urgent need for gene functional characterization, especially after the cassava full genome was sequenced. Methods Fully expanded leaves from in vitro plantlets of Manihot esculenta were used to optimize the concentrations of cellulase R-10 and macerozyme R-10 for obtaining protoplasts with the highest yield and viability. Then, the optimum conditions (PEG4000 concentration and transfection time) were determined for cassava protoplast transient gene expression. In addition, the reliability of the established protocol was confirmed for subcellular protein localization. Results In this work we optimized the main influencing factors and developed an efficient mesophyll protoplast isolation and PEG-mediated transient gene expression in cassava. The suitable enzyme digestion system was established with the combination of 1.6% cellulase R-10 and 0.8% macerozyme R-10 for 16 h of digestion in the dark at 25 °C, resulting in the high yield (4.4 × 107 protoplasts/g FW) and vitality (92.6%) of mesophyll protoplasts. The maximum transfection efficiency (70.8%) was obtained with the incubation of the protoplasts/vector DNA mixture with 25% PEG4000 for 10 min. We validated the applicability of the system for studying the subcellular localization of MeSTP7 (an H+/monosaccharide cotransporter) with our transient expression protocol and a heterologous Arabidopsis transient gene expression system. Conclusion We optimized the main influencing factors and developed an efficient mesophyll protoplast isolation and transient gene expression in cassava, which will facilitate large-scale characterization of genes and pathways in cassava.


Background
Transient gene expression in plant protoplasts is a powerful technique for studying subcellular localization of proteins, gene and promoter activities, protein-protein interactions, and signal transduction [1][2][3][4][5][6][7]. Compared with the stable gene expression in transgenic plants, transient gene expression represents a fast, convenient and efficient alternative system with higher expression levels [8][9][10]. Moreover, the transient gene expression assay enable the high-throughput analysis of gene functions while the stable gene expression is relatively expensive and timeconsuming thus limiting the utilization of this technique for large-scale analyses of plant genes [8][9][10]. Although different transfection techniques such as polyethylene glycol (PEG)-mediated, electroporation and microinjection have been developed to deliver recombinant DNA plasmids into protoplasts, utilization of the PEG-mediated approach has high transformation efficiency, so it is widely applied in molecular and cellular studies for both model plants and non-model plants [4,7,8,[11][12][13][14][15][16][17][18][19][20].
Cassava (Manihot esculenta Crantz) is a perennial and woody shrub of the Euphorbiaceae cultivated in tropical and subtropical regions for its starchy storage roots [21,22]. These roots have been used as the sources for dietary carbohydrate, starch processing and potential biofuel production. Since that cassava is a staple crop for approximately 800 million people in developing regions of the tropics [23], many groups have been intensively performing research on cassava molecular breeding which relies on the identification of agronomically important genes and pathways. Recently, the cassava sequencing data set has been publicly released [24] and the determination of gene function has been a major goal of cassava molecular biology in genomic and post-genomic era, thus it will necessitate efficient high-throughput transient gene expression for gene function analysis.
Stable gene transfer can be developed for gene function studies at whole plant level. Much progress has been made in cassava genetic transformation of both Agrobacterium tumefaciens-mediated and biolisticmediated system [23,[25][26][27]. However, the transformation efficiency is relatively low and genotypic dependent [28], and the process requires well-trained tissue culture specialists, is lengthy and difficult to repeat [29,30]. Hence, the development of an efficient transient gene expression system that could be exploited to characterize gene functions and investigate molecular processes in cassava is required.
Due to the fact that cassava is recalcitrant for plant regeneration from protoplasts, only two reports about the protoplast culture and regeneration of cassava have been published. Shahin and Shephard reported shoot regeneration from the isolated mesophyll protoplasts of cassava [31] and Sofiari et al. reported the plant regeneration from protoplasts isolated from friable embryogenic callus of cassava [32]. Anthony et al. developed a protocol for the culture of cassava leaf protoplasts but plant regeneration from protoplasts had not achieved [33]. To date, transient gene expression in cassava protoplasts has not been developed. Here we report an effective and reliable mesophyll protoplast isolation and PEGmediated transient gene expression in cassava. The protocol has been successfully used for subcellular localization of a sugar transporter, MeSTP7 protein.

Materials
In vitro plantlets of Manihot esculenta cv. South China 8 (SC8) were used in this study. Cultures were kept on 1/ 2MS (Murashige and Skoog) medium (supplemented with 2% sucrose and 0.8% agarose, pH5.8) at 28°C, under lighting with a cycle of 12 h/8 h (light/darkness) for 6-10 weeks to obtain fully expanded leaves.
The digested tissues were filtered through a 0.75-mm nylon sieve, collected by centrifugation (80 g for 3 min) and suspended in 20 mL pre-cooled W5 solution (154 mM NaCl, 125 mM CaCl 2 , 5 mM KCl and 2 mM MES, pH 5.8). The protoplast pellet was purified twice in W5 solution by repeated resuspension and centrifugation (80 g for 3 min).
The resulting protoplasts were resuspended in 2 mL W5 solution, placed on the ice for 30 min and counted under a microscope equipped with a hemocytometer. The viability of protoplasts was measured with 0.2% fluorescein diacetate (FDA) staining and determined as follows: protoplast viability (%) = (fluorescent protoplast number in view/protoplast total number in view) × 100%.
The transfected protoplasts were centrifuged (80 g, 2 min) twice, then resuspended with 300-500 μl WI solution (4 mM MES, 9% mannitol, 20 mM KCl, pH 5.8) and incubated overnight at room temperature in dark for the induction of protein expression. Transformation efficiency was detected under fluorescence microscopy and the expression of GFP tag was observed by confocal laser scanning microscope. Transformation efficiency was calculated as follows: transformation efficiency (%) = (bright green fluorescent protoplast number in view/ total protoplast number in view).

Plasmid construction and subcellular localization of MeSTP7 protein
Total RNA was extracted from cassava leaves with RNAplant Plus (Tiangen biotech, Beijing, China) and cDNA was synthesized using the PrimeScript RT reagent Kit with gDNA Eraser (TaKaRa, Dalian, China) following the manufacturer's instructions.
The specific primers with Xho I and Spe I restriction enzyme sites (bold) introduced to their ends respectively, STP7-F (5'-CCGCTCGAGATGCCTGCAGGAGGTTT-3') and STP7-R (5'-GGACTAGTGAAAACTGGGTAA-CAGGATCAA-3'), were designed according to the conserved region of MeSTP7 (GenBank accession number: Manes.03G180400). The gene of interest was amplified using Phusion DNA Polymerase under the following conditions: 98°C for 30 s followed by 35 cycles of amplification (98°C for 10 s, 57°C for 30 s, 72°C for 45 s) and a 10-min final extension at 72°C.
The PCR products were confirmed by sequencing and inserted between the 35S promoter and the GFP of the expression vector pA7-GFP, designated as pA7-MeSTP7-GFP (Fig. 1). The plasmid construct was propagated in competent Escherichia coli Trans-T1 (TransGen Biotech, Beijing, China). Subcellular localizations of MeSTP7 protein in cassava and Arabidopsis mesophyll protoplasts were carried out according to our transformation procedure described above and the method established by Yoo et al [3], respectively.

Isolation of protoplasts from cassava leaves
Different enzyme combinations significantly influenced the yield and viability of cassava protoplasts. Protoplast yield increased with increasing concentrations of cellulase R-10 and macerozyme R-10, but protoplast viability decreased with increasing enzyme concentrations. It was found that the highest production of protoplasts (5.1 × 10 7 protoplasts/ g FW) was obtained with 2.4% cellulose and 1.2% macrerozyme, with the lowest viability of protoplasts (83.8%). For obtaining the highest number viable protoplasts for the subsequent transfection experiments, the optimum combination of enzymes for cassava protoplast isolation was determined as 1.6% cellulase + 0.8% macerozyme. This combination gave the yield of 4.4 × 10 7 protoplasts/g FW and the viability of 92.6%. The enzyme solution turned green, indicating the release of mesophyll protoplasts. Leaf mesophyll protoplasts were uniformly spherical, and containing single intact nucleus and green chloroplasts. (Table 1, Fig. 2).

Establishment of cassava mesophyll protoplast expression system
In this study, the effects of PEG4000 concentration and transfection duration on cassava protoplast transformation efficiency was investigated with the transient expression vector pA7-GFP. The results showed that when the transfection time was 10 min, the transfection efficiency increased along the increasing concentration of PEG4000 and reached the maximum level (70.8%) at the concentration of 25% (Figs. 3 and 4a). When the concentration of PEG4000 was 25%, transfection efficiency increased with increases in the transfection time until it peaked at 10 min (Figs. 3 and 4b). Therefore, factors affecting transfection efficiency were optimized: 25% of PEG4000 concentration and 10 min of transfection time. In addition, stability of transient expression of GFP has been verified as GFP signal could be detected 8-30 h after transformation.

Subcellular localization of GFP-fused MeSTP1 protein in cassava mesophyll protoplasts
To test the feasibility of the protoplast transient expression system for the subcellular localization of protein in cassava cells, the transient expression of GFP-fused MeSTP7 protein in cassava mesophyll protoplasts was performed. Twelve-eighteen hours after transfection, brilliant green fluorescence distributed in intracellular compartments was observed. MeSTP7:GFP was specifically detected in cell membrane, but empty vector control (GFP) was distributed throughout the nucleus and cytoplasm (Fig. 5). Similar pattern of MeSTP7 subcellular localization in Arabidopsis mesophyll protoplasts validated the authenticity of the results described above (Fig. 6).

Discussion
A reliable protoplast isolation with high yield and viability is prerequisite for the success of transient gene expression. Source tissue, the type and concentration of enzymes are critical factors affecting the release of protoplasts [3,35,36]. Leaves from field or greenhousegrown plants usually have a thick cuticle and recalcitrant cell walls, which are likely to be the major limiting factors for protoplast isolation [12]. In cassava, Shahin et al. [31] used well expanded leaves which were collected from preconditioned plants (in growth chamber) and then incubated on a solution supplemented with naphthaleneacetic acid (NAA) and benzylaminopurine (BAP) as protoplast source. Sofiari et al. [32] isolated cassava protoplasts from friable embryogenic callus (FEC) and from suspensions derived from FEC. Anthony et al. [33] took juvenile leaves from axenic shoot cultures of cassava as protoplast source. For convenience sake, we chose the leaves from in vitro propagated cassava plants as the protoplast source.
For a given species and source tissue, suitable enzymolysis time and enzymolysis composition are vital for protoplast isolation. Anthony et al. [33] had established an improved protocol for cassava protoplast isolation with an enzyme mixture for 16 h. Referring to the method described by Anthony et al. [33], optimization of the enzymolysis composition was conducted in this study. The combinations of cellulase and macerozyme which have been commonly used for protoplast isolation, were evaluated to obtain protoplasts with the highest yield and viability. As the concentration of total enzyme in the mixture increased, protoplast yield seemed to increase but protoplast viability to decrease. High levels of cellulose R-10 and macrerozyme R-10 increased protoplast yield but decreased the viability of protoplasts, possibly due to the influence of enzymes on the integrity of membrane and the physiological activities of protoplasts [37,38]. The highest yield (5.1 × 10 7 protoplasts/ g FW) and the lowest viability of protoplasts (83.8%) were recorded with the 2.4% cellulose and 1.2% macrerozyme, which was the most concentrated enzyme mixture in our study. This lowest viability of protoplasts may not be suitable for subsequent experiments. However, the lowest concentrations of enzymes in the isolation mixture gave a lower yield (1.8 × 10 7 protoplasts/g FW) but with a slightly higher viability of protoplasts (93.6%). The enzyme mixture containing 1.6% Cellulase R10 and 0.8% macerozyme R10 was the best mixture in terms of both yield (4.4 × 10 7 protoplasts/g FW) and viability (92.6%) of protoplasts. The maximum yield and viability of leaf mesophyll protoplasts achieved in our study were substantially higher than the results previously reported [28,30]. The reason possibly attributed to the differences in tissue source used [31,32], and/or the type and concentration of enzymes [32,33]. In addition, the high yield (4.4 × 10 7 protoplasts/g FW) and vitality (92.6%) of cassava mesophyll protoplasts obtained in our study were comparable to the values reported for Populus (1 × 10 7 protoplasts/g FW, more than 90%) [13], Brachypodium distachyon (1.7 × 10 7 protoplasts/g FW, more than 90%) [19], cucumber (6-7 × 10 7 protoplasts/g FW, about 90%) [15]. The high yield and viability allowed the application of this system for the subsequent transfection experiment.
Although heterologous transient gene expression systems with Arabidopsis and tobacco protoplasts are commonly adopted for studying gene function in non-model plants such as cassava, these heterologous systems may display aberrant results due to a foreign genetic background. In this study, we established an efficient cassava protoplast transfection system by optimizing PEG4000 concentration and transfection time. This protocol will be an extraordinarily valuable tool for cassava gene functional study. In addition, transfection efficiency over 50% is generally required for obtaining reliable and reproducible results with the protoplast system [3]. The highest transfection efficiency (70.84%) with strong GFP signal was obtained with 25% of PEG4000 concentration for 10 min of transfection time. This result demonstrated that the transfection efficiency of cassava protoplasts was sufficiently high to be used for gene functional analysis in the future.
Using the optimized protocol that we developed, subcellular localization of MeSTP7 was investigated. Cassava MeSTP7 is a homologous gene of STP1 (SUGAR TRANSPORTER PROTEIN 1) in Arabidopsis thaliana and Nicotiana tabacum, which encodes an H + /monosaccharide cotransporter containing 12 putative transmembrane domains and is a putative plasma membrane protein [39][40][41]. Our results showed that MeSTP7 have been likely localized to the cassava plasma membrane. Similar to the localization in  cassava protoplasts, the MeSTP7 protein was distributed in Arabidopsis plasma membrane using a heterologous transient gene expression system. This confirmed that the cassava mesophyll protoplast expression is a reliable procedure for subcellular protein localization.

Conclusions
In summary, a highly efficient protocol for cassava mesophyll protoplast isolation and PEG-mediated transient expression was obtained through optimization of the type and concentration of enzymes for protoplast preparation, and PEG4000 concentration and treatment time for transfection. The method reported here can thus be very useful for molecular and cellular studies in cassava, especially combined with genetics, genomics, transcriptomics and proteomics. It is noteworthy that the simple incubation buffers described in this article do not support division of mesophyll protoplasts. Therefore, the system will not  appropriate to evaluate the expression of cassava genes involved in, e.g., cell cycle, plant dedifferentiation, or genes that require long-term follow-up (several cycles of cell division). The protoplast regeneration protocols developed by Shahin and Shephard [31], Sofiari et al [32] and Anthony et al [33] can be used for this purpose.