Experimental Design & Results

Introduction

---

Our experiments aimed to assess whether TaHsp70 is a heat-inducible promoter. Furthermore, we sought to recapitulate this activity in a wheat-like system such that in downstream applications, our findings can be directly applied when expressing our enzymes of interest: ACC deaminase, SBPase, and choline monooxygenase, in Triticum aestivum wheat.

Part of emulating such activity involves individually testing the expression of each of these enzymes under the control of the TaHsp70 promoter and determining how this model will respond to increased heat conditions. Enzyme expression was tested in plant-like systems called protoplasts which are plant cells with the cell walls digested. The cell walls of these cells are digested by specific enzymes which allow our constructs of interest to successfully be expressed in the plant cell. Plant protoplasts are efficient cell-based systems which allow quick and simultaneous transient analysis of multiple genes^[1].

The wheat TaHsp70 promoter was chosen after an extensive search for endogenous heat-related wheat promoters. Due to it having the highest activity in comparison to other known wheat heat-inducible promoters^[2], TaHsp70 served as a good candidate for testing in the Wet-Lab. Furthermore the promoter was tested in two of our constructs in order to upregulate the expression of the genes coding the following enzymes: AAC Deaminase and SBPase. This was done via heat shocking transfected protoplasts at a temperature of 37°C for 2.5 hours.

As stated, accomplishing this goal relies on two components: creating an artificial wheat-like system and testing our enzymes in controlled conditions within this system. The designed experiments seek to validate the functionality of TaHsp70 through heat-induced expression of our gene circuits. Our experiments were aimed at building our constructs in-vitro, as well as developing and testing them within a plant sandbox as a model environment.

Building Constructs

---

As a design consideration, a control construct was developed to validate whether any significant differences in observed fluorescence were due to the regulated activity of TaHsp70. This control construct was designed with a constitutive plant promoter, CaMV35S. This promoter would provide a baseline fluorescence intensity output for comparison of fluorescence intensities observed under heat shock induced expression. The chosen fluorescent reporter to establish this baseline was green fluorescent protein (GFP). Subsequently, this allows for normalization of fluorescence values from other reporters against GFP for analysis. Furthermore, GFP was also used to control for differences in transfection efficiency when compared to the other reporters through the process of normalization.

In tandem, the ACC deaminase and SBPase gene circuits were developed under the control of TaHsp70, and linked to separate fluorescent reporters without spectral overlap: mTagBFP and iRFP respectively. The addition of fluorescent reporters co-expressed with our enzymes is a simple readout for the successful expression of our enzymes under heat conditions. These constructs were assembled using Golden Gate Assembly.

Golden Gate was the chosen assembly method due to its ability to easily piece multiple DNA fragments together. Given the repeated use of the same heat shock promoter and terminator, this cloning pipeline significantly increased the speed of our Wet-Lab cloning activities.

We selected clones of interest using antibiotic resistance cassettes. Single colonies were expanded and assessed using Sanger sequencing. Troubleshooting was performed throughout this process to confirm plasmid sizes by gel electrophoresis following PCR amplifications and DNA ligations.

In-silico Genetic Circuit Golden Gate Assemblies

Demonstrating Heat Inducibility

---

As a proof-of-concept that the circuits are heat-inducible in the endogenous wheat system, we developed a heat treatment experimental plan to validate protein expression. Based on literature^[2], the optimal temperature for induction of TaHsp70 is 36°C. Protoplasts transfected with our constructs were heat shocked for 2 hours at a range above and below the optimal temperature: 30°C, 37°C, 45°C, and room temperature as a control. This occurred after the 16-hour incubation for transfection to allow for gene expression. 100 µL aliquots of transfected protoplasts were added in duplicates to 96-well flat-bottom plates (one for each heat treatment) and covered in aluminum foil during heat shock incubations. Protoplasts constitutively expressing GFP were used as a positive control of successful transfection, while untransfected protoplasts were used as a negative control.

Following heat shock, the fluorescence intensity (RFUs) for each fluorescent reporter corresponding to each gene circuit was measured at the respective excitation and emission wavelengths (475 and 545 nm for GFP; 660 and 710 nm for iRFP; 399 and 456 nm for BFP). The fluorescence data was then normalized against the baseline fluorescence generated from the constitutive CaMV35S promoter.

As observed in the three graphs below, fluorescence values for BFP and iRFP were only non-negligible when under the near-optimal heat conditions (37°C), and at 45°C fluorescent values were still present, but slightly lower.

Temperature treatments at 30°C and room temperature showed no fluorescence in BFP (top graph) nor iRFP (middle graph) fluorescent ranges for both SBPase and ACCD.

As expected, SBPase demonstrated no BFP nor GFP fluorescence, and ACCD demonstrated no iRFP nor GFP fluorescence, for any temperature treatment, while GFP, our positive control, demonstrated fluorescence at all temperatures. GFP values were not normalized since BFP and iRFP values were being normalized against it.

In subsequent experiments, we would like to add more temperature treatments at smaller intervals, with higher temperatures (up to 55°C) to more closely simulate the temperatures reached during heat waves (some of which are higher than 45°C, our highest temperature tested in this experiment).

We wanted to investigate how much of the lack of fluorescent data could be explained by protoplasts that lost viability during the heat shock; therefore, we measured viability before and after heat shock. This was done through adding 1 µL of 1% Evans Blue Dye to 25 µL of protoplasts and counting viable (non-blue) protoplasts on the hemocytometer. Viability percentage was calculated by the number of viable protoplasts over total. Previous studies^[3] showed no viability of protoplasts at 55°C, so we wanted to explore if this was also the case at lower temperatures. As shown below, protoplast viability did not vary considerably between temperatures, with 45°C showing the most decreased viability, potentially explaining the decreased fluorescence data seen at this temperature. Fluorescence did not decrease to the degree of protoplasts losing viability completely at this temperature; thus we hypothesize that our fluorescent proteins are remaining inside non-viable cells for a period of time before they are completely degraded. This means that some of the fluorescense can be attributed to non-viable protoplasts.

GFP fluorescence was also observable by fluorescence microscopy. The morphology of protoplasts is more irregular than the consistent spherical shape observed during protoplast purification and isolation. This change is likely due to handling and pipetting the protoplasts repeatedly, shearing them and modifying their shape. Some of these morphological extremes are merely pieces of protoplasts that are still expressing GFP: this supports our theory that non-viable or sheared protoplasts are still expressing fluorescent proteins for a period of time before their complete degradation, affecting fluorescence values.

Working with Protoplasts

---

Initially, we proposed the use of cell-free systems. However, various experts we talked to also expressed concern for a cell-free chassis. For instance, it was made clear that for optimal expression conditions to take place in cell-free systems, they require delicately prepared cell extracts. Given the fragility of chloroplast machinery suspended in solution, it was rationalized that under heat stress, reactive oxygen species would not be controlled for in-vitro, potentially disrupting protein synthesis processes. Furthermore, René Ickerman from the iGEM Plant Engineering committee mentioned that heat inducibility would not be easily feasible in a cell-free system since it is effectively a snapshot of cellular material that does not respond normally to stimuli.

Other iGEM teams like 2021 Marburg have previously developed chloroplast cell-free systems for genetic engineering work for Triticum aestivum (wheat). However, we could not use a chloroplast cell-free system for our project since not all of the enzymes we wanted to use were naturally expressed in the chloroplast genome and localized in the chloroplast lumen. In natural plants, the enzyme choline monooxygenase is found in the cytoplasm of cells and ACC deaminase is an enzyme found in bacteria associated with plants.

We also proposed the idea of agroinfiltration since it outputs the meaningful readout of a whole-plant effect. However, since our system is transiently expressed, we were told by Dr. Guus Bakkeren that it may not be strong enough to see its results on the whole plant level. In addition, the reagents we would need to purchase for agrobacterium-mediated transformation were higher in number, increasing shipping times and costs. Agroinfiltration also involved tissue culture, which would be time-consuming in addition to the increased costs. Even though we had support from Dr. Guus Bakkeren in providing agrobacterium strains, we decided against this option as well. Effectively, the cumulative technical expertise provided by our Integrated Human Practices interviews allowed us to confidently pivot in a direction away from chloroplast cell-free systems and towards protoplasts due to their ease of culturing, reduced cost of maintenance, and simple techniques for transfection. Apart from René Ickerman’s insights, Dr. Liang Song provided advice and support for our protoplast isolation, and Dr. Guus Bakkeren provided a protocol from a previous graduate student as well as spring wheat seeds to begin our plant growth.

Our team set about to use protoplasts for our proof of concept experiment to show that our plasmids with heat-inducible promoters could be expressed. Protoplasts are plant cells with their cell walls removed, that have much better transformation success than plant cells which retain their cell walls. We also decided to use wheat protoplasts because the heat-inducible promoter in our genetic circuits required a wheat-specific transcription factor, preventing us from using protoplasts from another model plant like Arabidopsis thaliana. We isolated our protoplasts from 7-10 day old seedling leaf tissue; this plant organ expresses our promoter’s transcription factor, TaHsf2Ab, and was convenient to extract from the early wheat.

Although the process of using protoplasts for plant transfection experiments is a fairly common practice, making wheat protoplasts is not well characterized in the literature^[4] and proved to be a very challenging protocol to optimize in the lab. Our team was able to generate viable protoplasts after several rounds of optimizing our protoplast isolation protocol (details can be found on Engineering Success page). Hopefully the development of this protocol could help other iGEM teams hoping to work with wheat plants in the future. For more specific documentation on optimization and common downfalls for other teams to build off of, please visit our Engineering Success and our Protocols pages.

We are fairly certain we are the first iGEM team to use wheat protoplasts, and one of the first teams to use plant protoplasts at all in an iGEM project.

Optimizing Protoplast Isolation

---

Protoplast isolations are tricky systems to work with and multiple rounds of optimization were done to ensure a high yield of cells.

The initial protocol used resulted in minimal success over the first two rounds of isolation. All cells were either sheared or lysed due to the delicate nature of protoplasts without a cell wall to protect them.

We obtained advice to use serological pipettes, whose edges are more rounded than micropipettes, to handle protoplasts at all times. If volumes were too small or serological pipettes limited, we were advised to cut the tips of p1000 micropipette tips to allow a wider area to decrease the pressure when pipetting. Incubation periods were shortened from the original 5 hours to about 3 hours as longer incubation times started lysing the cells due to overdigestion. Furthermore, the cell wall enzyme solution that was being used was frozen and not heat treated prior to isolation which was a leading cause of low cell counts. Initial rounds of optimization included cutting the wheat leaves in mannitol which we later realized resulted in loss of yield due to protoplasts lysing from the high osmotic pressure. This was followed by the addition of concentrated sucrose solutions to increase protoplast yield which we again realized was not significant.

After three unsuccessful rounds, we switched protocols to the current one documented in our Protocols page which yielded a high concentration of viable cells. This protocol required fresh preparation of enzyme solution which was heat treated prior to use. The leaves were cut into a dry petri dish instead of in mannitol, and we ensured that they were immediately transferred into the enzyme solution. Cell counts were checked hourly to make sure high yields were obtained and the rest of the protocol was carried out when protoplasts covered more than 60% of view under the light microscope during the incubations with cell wall enzyme solution to prevent overdigestion, to ensure maximum efficiency.

Viability Testing of Protoplasts

---

To validate that live protoplasts were successfully isolated from wheat seedlings, we used a hemocytometer to periodically monitor aliquots of the protoplast extract solution. This was done under a light microscope at hourly intervals during isolations, and again at the end of the procedure (the full procedure can be found on our Protocols page).

Live protoplasts appear as spherical green cells approximately 100 µm in diameter (Figure 1), whereas lysed protoplasts appear as irregularly-shaped cells or clusters of small green chloroplasts (Figures 2 and 3). As a further validation measure to ensure that protoplasts were still alive, we added 1 µL of 1% Evan's Blue dye to 25 µL aliquots in the hemocytometer, which stains dead protoplasts blue. Viability could be confirmed if green protoplast cells still appeared in the hemocytometer with Evan’s Blue dye added.

Storage Condition Optimization

As no long-term storage options currently exist for wheat protoplasts in the literature, we tested our own storage method based off a paper where F. brasiliense protoplasts were stored frozen at − 80°C for up to two years and still showed viability[5]. Being able to store protoplasts would be very useful because obtaining these systems is time consuming and not always reliably guaranteed. Finding a way to store wheat protoplasts allows us to use them in the future without worrying about yield and the possibility of not obtaining sufficient amounts for downstream experiments. We attempted to store our protoplasts by adding DMSO to 7% total concentration 2 mL cryotubes and cooling them in the -80°C freezer. The concentration of protoplasts frozen was twice the concentration (2.5 x 10⁶) needed for transfection, as we expected ~50% to lose viability during freezing or thawing. Protoplasts were thawed on ice, pelleted, and the supernatant removed. Subsequent washes were done with W5 to remove DMSO, and the protoplasts were finally resuspended in MMG. When the protoplasts were checked for viability using the hemocytometer the next day, we discovered that almost all of them were dead. We hypothesized that the protoplasts were killed by flash-freezing since we had no method to control the rate of temperature change after putting them in the freezer. Given further time and resources, we could potentially have improved the efficiency of this storage method by purchasing a Thermo Scientific™ Mr. Frosty™ Freezing Container, which controls the rate of cooling to approximately -1 °C/min.

After this initial testing at -80°C, we hypothesized that storage at higher temperatures could increase viability after storage in DMSO. We used the SBPase transfection condition as our sample to test longer-term viability testing. The graph below includes the temperatures before treatment, after 2-hour treatment, and viability after 24-hour storage in room temperature, 4°C and -80°C. Viability remained nearly the same after 24 hours and 48 hours at all conditions. The 4°C treatment after 24 hours showed an increased amount of viability, which was likely a bias error from counting on the hemocytometer.

Future Directions

---

Validation of Construct Function

After confirming that our enzymes were properly expressed from our constructs into protoplasts, we would need to validate that the enzymes are working as proposed. The outcome we were hoping for is increased wheat crop yield and reduced loss of crop viability after heat; however, we do not have the required resources and permission within an iGEM context to conduct field trials. Hence, realistically for an iGEM team, we aimed to validate that our enzymes were working correctly at the molecular and cellular level. For such characterization, we aimed to measure transcript levels of gene pathways that our enzymes are involved in, or assay the enzymatic activity of the expressed enzymes. The following sections below outline such validation experiments.

SBPase & the CBB Cycle

We were able to quantify the expression of the SBPase protein through fluorescence measurements of the co-expressed iRFP product, which by design would be expressed in equimolar amounts as the SBPase product. This is due to the mechanism of ribosome skipping induced by the P2A linker^[6]. 2A self-cleaving peptides (the class to which P2A belongs), when tagged to proteins of interest, trigger ribosomal skipping during translation. Using P2A is a way to generate multiple proteins simultaneously from the same transcript, without needing to place each protein under a separate transcriptional unit. Therefore, our design should in theory ensure equal amounts of SBPase and iRFP production, which allows for iRFP levels to serve as an accurate measurement of SBPase level.

Quantification of SBPase protein function is more difficult, but we aimed to follow the total SBPase activity assay described by Driever and colleagues^[7]. In this protocol, total protein extract from snap-frozen leaf tissue was incubated with SBP in reaction buffer and the produced phosphate from the SBPase catalyzed reaction was measured through spectrophotometry with the aid of a phosphate interacting dye (BIOMOL Green). As we are using protoplasts for our expression system, total protein extraction would likely require fewer steps as no separation of protein from cell wall and tissue is required.

Ethylene Measurement with Gas Chromatography

Ideally, we hoped to quantify the activity of ACC deaminase during heat shock by measuring the quantity of ethylene produced relative to a negative control. We expect that plant cells expressing ACC deaminase would have lower ethylene production during heat stress relative to control plant cells due to ACC deaminase breaking down the precursor to produce ethylene. Our plan for quantifying the effect of heat stress on ethylene production involved taking measurements of ethylene production before and after heat shock exposure for ACC deaminase expressing plant cells and control plant cells. Upon purging aliquots of the individual samples with a carrier gas like nitrogen, produced ethylene gas can be collected from protoplasts. Aliquots would be taken since a given sample would need to be measured multiple times such as before and after a heat shock treatment, and the purging process is destructive, so the entire sample cannot be used for measurement. These gas samples would then be run on a Gas Chromatography - Mass Spectrometry (GC-MS) instrument, which would ultimately allow us to infer the relative content of gaseous ethylene in the sample aliquots.

Downstream Ethylene Regulation Studies

In addition, it would be of interest to investigate downstream targets of the ethylene signaling pathway in wheat. Current literature has pointed towards specific gene targets belonging to the ERF Family (Ethylene Response Factors)^[8]. These factors bind regulatory elements in pathways associated with abiotic stress management. This may open doors for our team to explore how drought stress regulation is tightly woven with salinity and high light tolerance and molecular mechanisms by which this regulation occurs (ie. cooperation occurring between hormones and ERFs, and their effect on expression of our proteins).

Choline Monooxygenase Construct In-Vitro

Due to time constraints, we were unable to complete building the choline monooxygenase construct in-vitro in the midst of optimizing the protocol for protoplast isolation and transfection. Similar to the other constructs developed in-vitro, this circuit would have also been fused to fluorescent reporter protein for characterization of function under heat inducibility. As described in our Project Design page, we utlimately aimed to show that the combinatorial expression of these three gene circuits would result in improvement in water retention and the biosynthesis of key proteins involved with growth in wheat plants.

Stable Protoplast Transfection and Achieving Transgenic Wheat

Downstream, we would have liked to transfect our constructs into Triticum aestivum through agrobacterium-mediated gene transfer transfection. This method allows for integration of genetic material into the genome. As its use is relatively more simple than other gene editing tools, our team selected this as the primary method to achieve transgenic wheat. Agrobacterium tumefacians is a strain that has been used widely to transfer DNA into plants for genetic engineering of whole plants. It accomplishes this through its activation of virulence genes that aid in moving its T-DNA (Transferred DNA) into plant cell nuclei^[9]. This natural “genetic engineer” would allow us to accomplish transfection without the need for additional genetic editing tools, increasing feasibility and ease of use.

Alternatively, there are other methods to achieve a transgenic wheat crop using CRISPR-Cas ribonucleoprotein complexes^[4]. Our team would have liked to compare how a CRISPR edited system compares to that of agrobacterium mediated genome editing. CRISPR would involve coexpression of our sgRNA and Cas protein and subsequent purification of the ribonucleoprotein^[10]. The transfection process for stably integrating constructs is highly similar to the process we used for transfection, the only notable difference being the addition of Cas ribonucleoproteins to the protoplast solution prior to addition of PEG4000 and subsequent PEG-mediated transfection. The Cas-9 ribonuceloproteins generate double stranded breaks at the target site at which the cotransfected construct may be integrated through homologous recombination. Transfected protoplasts may then be grown in Murashige-Skoog media to generate calli which may then be differentiated through tissue culture to generate full plants^[11]. Subsequent qPCR relative to a gene of known copy number may be used to determine the number of copies of the construct integrated for individual plants propagated from singular transfected protoplasts.

Measuring mRNA Transcript Levels of Expressed Gene Circuits

Down the line, it was the hope that additional characterization of our gene circuits would take place at a protein level to add characterization to potential gene dosage studies in-vitro. Ideally, we would have measured a baseline steady state of protein expression for ACC deaminase, SBPase, and choline monooxygenase through mRNA transcript quantitation methods (ex: RT-qPCR, Northern Blot). The relative amount of mRNA in a cell can approximate the relative amounts of protein in a cell. Understanding of this baseline would have helped elucidate mechanisms of transcription and patterns observed under high heat conditions for our three enzymes.

Implemented Optimization Procedures

Additional optimization techniques that were included to help improve the ease at which we are able to assemble our genetic circuits in-vitro are as follows:

1) Increasing cell density of E. coli hosts in media.

We supplemented our LB media with yeast extract media to increase the cell density in liquid cultures. The goal was to increase plasmid extraction yields in preparation for protoplast transfections. However, due to time constraints, we were unable to generate quantifiable measurements that evaluated the success of our method over traditional protocols for culturing.

2) Increasing plasmid extraction yield.

As protoplast transfections require a high concentration of DNA, we utilised an ethanol precipitation method in hopes of increasing our final concentration compared to the yield achieved from standard plasmid extraction kits. From DNA spectrophotometry measurements, our method of ethanol precipitation reliably produced higher DNA concentrations.