A high-throughput, automated, tobacco protoplast production and transformation methodology is described. The robotic system enables massively parallel gene expression and discovery in the model BY-2 system that should be translatable to non-model crops.
Over the last decade there has been a resurgence in the use of plant protoplasts that range from model species to crop species, for analysis of signal transduction pathways, transcriptional regulatory networks, gene expression, genome-editing, and gene-silencing. Furthermore, significant progress has been made in the regeneration of plants from protoplasts, which has generated even more interest in the use of these systems for plant genomics. In this work, a protocol has been developed for automation of protoplast isolation and transformation from a 'Bright Yellow' 2 (BY-2) tobacco suspension culture using a robotic platform. The transformation procedures were validated using an orange fluorescent protein (OFP) reporter gene (pporRFP) under the control of the Cauliflower mosaic virus 35S promoter (35S). OFP expression in protoplasts was confirmed by epifluorescence microscopy. Analyses also included protoplast production efficiency methods using propidium iodide. Finally, low-cost food-grade enzymes were used for the protoplast isolation procedure, circumventing the need for lab-grade enzymes that are cost-prohibitive in high-throughput automated protoplast isolation and analysis. Based on the protocol developed in this work, the complete procedure from protoplast isolation to transformation can be conducted in under 4 hr, without any input from the operator. While the protocol developed in this work was validated with the BY-2 cell culture, the procedures and methods should be translatable to any plant suspension culture/protoplast system, which should enable acceleration of crop genomics research.
In recent years there has been significant impetus placed on the design of transgenic crops to overcome various diseases1, endow herbicide resistance2, confer drought3,4 and salt tolerance5, prevent herbivory6, increase biomass yield7, and decrease cell wall recalcitrance8. This trend has been aided by the development of new molecular tools for generating transgenic plants, including genome-editing using CRISPR and TALENs9, and gene silencing through dsRNA10, miRNA11, and siRNA12. While these technologies have simplified the generation of transgenic plants, they have also created a bottleneck, where the sheer number of transgenic plants generated cannot be screened using traditional systems that rely on plant regeneration. Related to this bottleneck, while silencing and genome-editing constructs can be rapidly inserted into plants, many of the targeted traits fail to produce the desired effect, which is often not discovered until plants are analyzed in the greenhouse. In this work, we have developed a method for rapid, automated, high-throughput screening of plant protoplasts, specifically to address the current bottleneck in early screening of large numbers of genome-editing and gene silencing targets.
The use of protoplasts, as opposed to intact plant cells, has several advantages for the development of an automated platform. First, protoplasts are isolated after digestion of the plant cell wall, and with this barrier no longer present, transformation efficiency is increased13. In intact plant cells there are only two well established methods for transformation, biolistics14 and Agrobacterium-mediated transformation15. Neither of these methods can be easily translated to liquid handling platforms, as biolistics requires specialized equipment for transformation, whereas Agrobacterium-mediated transformation requires co-culture and subsequent removal of the bacteria. Neither are amenable for high throughput methods. In the case of protoplasts, transformation is routinely conducted using polyethylene glycol (PEG)-mediated transfection16, which only requires several solution exchanges, and is ideally suited for liquid handling platforms. Second, protoplasts, by definition, are single-cell cultures, and thus the problems associated with clumping and chain formation in plant cell cultures, are not observed in protoplasts. In terms of rapid screening using a plate-based spectrophotometer, clumping of cells, or cells in multiple planes will lead to difficulty in acquiring consistent measurements. Since protoplasts are also denser than their culture media, they sediment to the bottom of wells, forming a monolayer, which is conducive for plate-based spectrophotometry. Finally, while plant cell suspension cultures are primarily derived from callus17, protoplasts can be harvested from a number of plant tissues, leading to the ability to identify tissue-specific expression. For example, the ability to analyze root- or leaf-specific expression of a gene can be very important to phenotype prediction. For these reasons, the protocols developed in this work were validated using protoplasts isolated from the widely-used tobacco (Nicotiana tabacum L.) 'Bright Yellow' 2 (BY-2) suspension culture.
The BY-2 suspension culture has been described as the "HeLa" cell of higher plants, owing to its ubiquitous use in molecular analysis of plant cells18. Recently, BY-2 cells have been used to study the effects of plant stressors19-22, intracellular protein localization23,24, and basic cell biology25-27 demonstrating the broad utility of these cultures in plant biology. An additional advantage of BY-2 cultures is the ability to synchronize the cultures with aphidicolin, which can lead to enhanced reproducibility for gene expression studies28. Furthermore, methods have been developed for the extraction of BY-2 protoplasts using low-cost enzymes29,30, as enzymes traditionally used for generating protoplasts are cost prohibitive for high-throughput systems. As such, the protocol described below has been validated using the BY-2 suspension culture, but it should be amendable to any plant cell suspension culture. Proof-of-concept experiments are performed using an orange fluorescence protein (OFP) reporter gene (pporRFP) from the hard coral Porites porites31 under the control of the CAMV 35S promoter.
1. Establishment of Suspension Cell Cultures
2. Protoplast Isolation
3. Establish Standard Curve for Cell Counts
4. Microscopic Analysis of Transformation
In the current study, the doubling rate of BY-2 varied from 14-18 hr dependent on the temperature at which the cultures were incubated, consistent with previous reports of a mean cell cycle length of 15 hr. With this doubling rate, a 1:100 starting inoculum was used to initiate cultures, leading to cultures with a packed cell volume (PCV) of 50% in 5-7 days. In the current protocol, in which cultures were grown in 200 ml of media, a PCV of 100 ml was generated in 7 days, which provided enough cells to fill 33 6-well plates. In terms of protoplast yield, digestion at the conditions specified in the protocol generated 4.70 x 105 ± 5.77 x 103 protoplasts per well, resulting in 2.82 x 106 protoplasts per 6-well plate (Figure 2A). Based on these data, 214 96-well plates (70 µl of protoplasts per well) could be loaded from a single 6-well plate digestion, with the potential for two 6-well plates to be digested simultaneously, utilizing the two plate heater/chiller stations. Thus, the maximum number of protoplasts that can be generated during a single three-hour digestion protocol is 5.64 x 106.
While the total number of protoplasts generated per digestion protocol provides one metric for the system, it is also necessary to evaluate the loss in protoplast concentration as the protoplasts are transferred through the different steps of the protocol. As such, the concentration of protoplasts was evaluated after each transfer step in the protocol. After the digestion protocol, 1.15 x 104 ± 1.35 x 102 protoplasts were transferred to each well of the 96 deep-well plates, if conducting the transformation protocol, or the 96-well fluorescent screening plate (Figure 2B). Once the transformation protocol was completed, 1.23 x 103 ± 4.66 x 101 protoplasts were transferred to the 96-well fluorescent screening plate (Figure 2C). In addition to the number of protoplasts transferred, it was important to determine the viability of the protoplasts to ensure that multiple pipetting stages did not damage the protoplasts.
Based on the results from the viability assay, 95.1 ±0.04% of protoplasts were viable after digestion, 92.1 ± 0.06% after transfer to the 96-well fluorescent screening plate, and 91.7 ± 16.67% after the transformation protocol. Despite multiple pipetting steps, there was no significant difference (p > 0.44) between any of the samples. In addition to this measure of viability, a protocol was developed for determining the concentration of protoplasts at any stage of the procedure by measuring the total number of protoplasts stained with PI. As shown in Figure 3, a standard curve was generated by measuring the fluorescence intensity for known concentrations of protoplasts from the system. The standard curve was a good fit (R2 = 0.967) with the total number of protoplasts spanning the range from 0 to 30,000 protoplasts per well. To validate the standard curve, the fluorescence intensity of samples with a known number of protoplasts (8,125 and 24,375 calculated using a hemocytometer) was obtained and the number of protoplasts calculated from the equation of the standard curve. Based on these results, the standard curve predicted 9,560 and 28,200 protoplasts in each well, a 15 and 14% error. Considering that there is also variability in measuring the cell number using a hemocytometer, this error was deemed acceptable.
In addition to the results of the cell number and viability, the transformation protocol was successful in generating transgenic cells that express OFP (Figure 2D–F). Unfortunately the protocol used resulted in low, ~2%, transformation efficiency, due to the large 16,028 bp binary plasmid used in this study and since the protocol was not optimized specifically for BY-2 protoplasts. Optimization of the transformation protocol, in terms of reagents and conditions specifically for BY-2 protoplasts, is expected to increase the transformation efficiency, as elucidated in the Discussion. In terms of this protocol, the transformation procedure was designed to demonstrate the proof-of-concept path from protoplast isolation to transformation, and was successful in achieving this goal.
After successfully validating all of the individual procedures, metrics were obtained for the timecourse of the protocol from protoplast isolation to transformation. As demonstrated in Figure 4, the major investment of time is spent during the digestion stage, in which 3 hr was required for complete digestion of the cell wall and release of the protoplasts. During this procedure, 18 loops are run between the plate shaker and plate heater/chiller 1 incubation station, with the plate mover moving the plate between the two stations. The transfer time between the plate shaker and plate heater/chiller 1 takes just 8.9 sec, and ensures that the majority of the digestion procedure is spent incubating and shaking. The total time from initiation of the procedure to complete loading of the enzymes by the multi-mode dispenser is 2.2 min. Overall, the complete protoplast isolation protocol is completed in 3 hr and 22.8 min. Based on these results, 88% of the protoplast isolation protocol is spent digesting. After completion of the protoplast isolation protocol, the transformation protocol is initiated, and completed within 27.5 min. Similar to the protoplast isolation protocol, 72% of the transformation procedure is spent incubating the reaction. The only other step in the protocol that has a significant time component (>10% of the total procedure time) is initialization of the protocol by the automated liquid handling platform, which accounts for 14% of the total time in the procedure. In this way, 86% of the protocol is spent in initialization and incubation, with only 14% of the time spent on pipetting and transfer steps. The complete duration of protoplast isolation and transformation was 3 hr, 50 min, and 53 sec, with no external input required from the operator.
Figure 1: Schematic of the robotic platform and configuration of the automated liquid handling platform during the protocol. (A) The robotic system is composed of 3 plate stacks/hotels that can be accessed by the plate mover. Hotel 1 is the designated lidding/delidding station, Hotel 2 is the 96-well fluorescent plate stack, and Hotel 3 is the 6-well plate stack. The system has two liquid handlers, an automated liquid handling platform, and a nozzle-based multi-mode dispenser. For heating/cooling the robotic platform has two plate heating/chilling units, along with a plate shaker for mixing the plate. Finally, the system has a monochromator-based plate reader for measuring fluorescence. All components are housed in a custom built Biosafety Cabinet. (B) Starting configuration of the automated liquid handling platform with nest 2 occupied by the reagent plate and the tip box placed on nest 8. (C) Configuration of automated liquid handling platform prior to calling an experimental protocol. A deep-well plate is located on nest 4, a 96-well fluorescent screening plate is located on nest 6, and the 6-well plate containing protoplasts is located on nest 5. Please click here to view a larger version of this figure.
Figure 2: Representative images of BY-2 protoplasts after each transfer and after transformation with the orange fluorescent protein reporter gene. (A–C) Micrographs of BY-2 protoplast density in 6-well plate, 96-well plate after transfer of 70 µl from the 6-well plate, and 96-well fluorescent screening plate post-transformation. (D) Brightfield micrograph of BY-2 protoplasts after transformation. (E) Fluorescent BY-2 protoplast expressing the OFP gene visualized using a Cy3/TRITC filter set. (F) Overlay of brightfield and fluorescent micrographs demonstrating the low efficiency of transformation. Scale bar = 50 µm. Please click here to view a larger version of this figure.
Figure 3: Standard curve for calculating the number of protoplasts. The number of protoplasts in a given well can be calculated by comparing the fluorescence intensity, in arbitrary units (AU), to the equation for the standard curve. Validation of the protoplast number was compared between the standard curve and hemocytometer at the point indicated on the graph, with 9,561 protoplasts identified from the plate reader, and 8,125 counted on the hemocytometer. Overall, an error of 15% was found between the two techniques. Please click here to view a larger version of this figure.
Figure 4: Gantt chart for the described procedure for the production of BY-2 protoplasts and genetic transformation. The majority of the protocol, ~3 hr, is spent on digestion of the BY-2 cells to generate the protoplasts. Loading of the enzyme solution and inactivation require only 12% of the overall procedure time. Similarly, incubation of the transformation reaction on the plate shaker accounts for 72% of the transformation protocol. Overall, the procedure from protoplast isolation until genetic transformation is achieved is under 4 hr (3 hr, 50 min, and 53 sec). Please click here to view a larger version of this figure.
Supplemental File: Automated protocols for protoplast isolation, cell counting, and PEG-mediated transformation. Complete protocols developed for use in the robotic platform to accomplish each of the tasks listed above are included in this file. Please click here to download this file.
The protocol described above has been successfully validated for protoplast isolation, enumeration, and transformation using the BY-2 tobacco suspension cell culture; however, the protocol could easily be extended to any plant suspension culture. At present, protoplast isolation and transformation has been achieved in numerous plants, including maize (Zea mays)10, carrot (Daucus carota)32, poplar (Populus euphratica)33, grape (Vitis vinifera)34, oil palm (Elaeis guineensis)35, lettuce (Lactuca sativa)36, mustard (Brassica juncea)37, rice (Oryza sativa)38, and switchgrass (Panicum virgatum)39,40. While variation in the culture and digestion conditions occur from species to species, in terms of the protocol developed in this work, replacing culture media or altering the conditions of digestion should not change the fundamental protocol. To adjust to different species, culture media and enzymes required for digestion are loaded by the operator prior to initialization of the protocol, thus swapping these components is trivial. As an added benefit, the protocol is amendable to screening multiple digestion conditions with the yield of protoplasts measured by the system. This is an essential component, as the use of low-cost enzymes are necessary for high-throughput applications, thus it is important to minimize the amount of enzymes used in digestion.
Similar to culture and digestion, modifications to the transformation protocol would be required when adopting the system for transformation of protoplasts isolated from different plant species. Considerable work has been conducted on optimization of PEG-mediated transformation in plant protoplasts16,35, with numerous factors affecting the efficiency of transformation. In the current work one limiting factor affecting the transformation efficiency was the large size of the binary test plasmid, 16,028 bp, which was used for the purposes of modeling a genome-editing plasmid size. In previous studies on PEG-mediated transformation of BY-2 protoplasts, 40-60% transformation has been achieved; however, these studies used small 5,000 bp plasmids41,42. The primary modifications to PEG-mediated transformation among species include the concentration of PEG and MgCl2, amount and concentration of DNA, and reaction duration. As the PEG and MgCl2 solutions are introduced into the system on the reagent plate, and the DNA is pre-loaded into the deep-well plate, these conditions can be easily modified in the protocol described above. Additionally, increasing or decreasing the incubation time, along with modification of mixing speeds, can be precisely controlled by the robotic platform. The current limitation of the transformation protocol is the lack of screening using the plate reader to detect the fluorescent reporter, due to the low efficiency of PEG-mediated transformation in BY-2 protoplasts. In other species, such as Arabidopsis thaliana and maize16, PEG-mediated transformation is 40-90% efficient. In such systems transformants could be rapidly characterized using the plate reader on the robot. Thus, it would be possible to identify wells with positive transformants, and also quantify the level of expression. For applications, such as promoter screening, this feature would increase the utility of the current protocol, and will be added in later iterations of the system.
In addition to optimization of the transformation efficiency for BY-2 protoplasts, several modifications may increase the efficiency and overall throughput of the system. Previous studies on automation of protoplast transformation allowed protoplasts to settle to the bottom of wells, prior to the addition of wash solutions43. In the current protocol, protoplasts were mixed on the plate shaker to prevent settling and keep the protoplasts in suspension using flat bottom well plates. Furthermore, during transformation, protoplasts were not allowed to settle in the deep-well plates, to reduce the exposure to PEG, which can be toxic over prolonged exposure. As such, the protoplasts were diluted ~10-fold with W5 solution when transferred from the deep-well plate to the 96-well fluorescent screening plate, resulting in 1.23 x 103 ± 4.66 x 101 protoplasts per well. In order to increase the number of protoplasts transferred after transformation, an incubation step could be added to allow the protoplasts to settle, and the supernatant removed followed by addition of a smaller volume of liquid BY-2 media. Additionally, if a plate-based centrifuge was added to the system, this step could be achieved more rapidly. Another modification that could add functionality to the system would be the use of a live/dead fluorescence assay to determine the viability of the isolated protoplasts prior to transformation. In the current protocol, fluorescein diacetate (FDA) was tested as a live dye; however, the BY-2 protoplast background fluorescence obscured the FDA signal. If a more stable viability dye was used, then protoplast viability could be determined as the ratio of red (PI) to green (Calcein AM, etc.) fluorescence, and standardized similar to the cell counting protocol. These modifications would add further functionality to the system, and will be included in future protocols.
With regards to the current protocol, there are several critical steps that must be followed to ensure the success of the protocol. First, the age/health of the BY-2 cultures is essential to guarantee that viable protoplasts can be isolated for subsequent steps. As explained in step 1.3, by transferring a log phase culture to a fresh flask at a 1:50 ratio of cells to fresh media, and allowing the culture to grow for 5-7 days, the maximum number of viable protoplasts can be isolated. Allowing the cultures to grow for longer or shorter durations using the specified inoculum will significantly reduce the protoplast yield, and cause failure of subsequent protocols. A second critical step is step 1.5, since the use of standard pipet tips, instead of the wide-bore pipet tips specified, will cause the tips to clog and lead to the incorrect volumes being transferred. With regard to the setup of the robotic system, the arrangement of the plates on the automated liquid handling platform, step 2.4, is critical, with a modified setup preventing the head of the liquid handler from reaching all wells on the 6-well plate. Due to the design of the automated liquid handling platform, the 6-well plate must be placed in nest 5, or protocols selected for this instrument will fail. Similarly, failure to correctly define the labware in the plate mover software, step 2.2, will cause the instrument to fail to grab, or drop the labware resulting in termination of the protocol. For the cell count protocol, the critical step involves the fixation of the cells with ethanol, step 3.2. If cells are not fixed prior to labeling with PI, only nonviable cells will be labeled, instead of the entire population of protoplasts, and it will not be possible to count the number of protoplasts isolated. The final critical step, step 4.6, involves selection of the fluorescent filters for microscopic analysis of the expression of the pporRFP fluorescent marker. The use of other "red" filter sets allows for background fluorescence from chlorophyll, significantly increasing the signal in all chloroplast containing protoplasts, preventing calculation of the transformation efficiency. Carefully adhering to the specifications in these critical steps guarantees the greatest opportunity for success and the generation of the most reproducible dataset.
The authors have nothing to disclose.
This research was supported by Advanced Research Projects Agency – Energy (ARPA-E) Award No. DE-AR0000313.
Orbitor RS Microplate mover | Thermo Scientific | ||
Bravo Liquid Handler | Agilent | ||
Synergy H1 Multi-mode Reader | BioTek | ||
MultiFlo FX Multi-mode Dispenser | BioTek | ||
Teleshake | Inheco | 3800048 | |
CPAC Ultraflat Heater/cooler | Inheco | 7000190 | |
Vworks Automation Software | Agilent | Software used to control and write protocols for Agilent Bravo | |
Momentum Software | Thermo Scientific | Task scheduling software for controlling Orbiter RS | |
Liquid Handling Control 2.17 Software | Biotek | Software used to control and write protocols for MultiFlo FX | |
IX81 Inverted Microscope | Olympus | ||
Zyla 3-Tap microscope camera | Andor | ||
ET-CY3/TRITC Filter Set | Chroma Technology Corp | 49004 | |
Rohament CL | AB Enzymes | sample bottle | low-cost cellulase |
Rohapect UF | AB Enzymes | sample bottle | low-cost pectinase |
Rohapect 10L | AB Enzymes | sample bottle | low-cost pectinase/arabinase |
Linsmaier & Skoog Basal Medium | Phytotechnology Laboratories | L689 | |
2,4 dichlorophenoxyacetic acid | Phytotechnology Laboratories | D295 | |
propidium iodide | Sigma Aldrich | P4170 | |
Poly (ethylene glycol) 4000 | Sigma Aldrich | 95904-250G-F | Formerly Fluka PEG |
Propidium Iodide | Fisher Scientific | 25535-16-4 | Acros Organics |
CaCl2 | Sigma Aldrich | C7902-1KG | |
Sodium Acetate | Fisher Scientific | BP333-500 | |
Mannitol | Sigma Aldrich | M1902-1KG | |
Sucrose | Fisher Scientific | S5-3 | |
KH2PO4 | Fisher Scientific | AC424205000 | |
KOH | Sigma Aldrich | P1767 | |
Gelzan CM | Sigma Aldrich | G1910-250G | |
6-well plate | Thermo Scientific | 103184 | |
96-well 1.2 ml deep well plate | Thermo Scientific | AB-0564 | |
96 well optical bottom plate | Thermo Scientific | 165305 | |
Finntip 1000 Wide bore Pipet tips | Thermo Scientific | 9405 163 | |
NaCl | Fisher Scientific | BP358-10 | |
KCl | Sigma Aldrich | P4504-1KG | |
MES | Fisher Scientific | AC17259-5000 | |
MgCl2 | Fisher Scientific | M33-500 |