PREreview of Agrochemical-responsive gene expression control systems for modulating plant development and metabolism

PREreview of “Agrochemical-responsive gene expression control systems for modulating plant development and metabolism” 

Overall Assessment 

Agrochemicals are strong candidates for incorporation into inducible systems in crops because they are formulated for efficient uptake and movement throughout tissues. In this work, Bull et al. focus on the agrochemical methoxyfenozide (MF), the active ingredient in a commercial insecticide. The authors increase the modularity of an existing MF-responsive ecdysone receptor (EcR) fusion protein by integrating Cas9- and Cre recombinase-based control systems. Bull et al. find that the initial fusions are leaky and lack a clean OFF state, but they demonstrate that leakiness can be decreased if a nuclear export sequence NES is included in the fusion proteins. In Arabidopsis stable lines, the authors show the systems’ ability to regulate expression of a fluorescent reporter. The authors also show that the Cas9-EcR system may be able to shift metabolic carbon flux when MF is applied, and that it enables some degree of ligand-dependent control of DELLA.

Overall, the evidence generally supports the authors’ conclusions. Plants treated with MF do show a significant increase in fluorescent output in transient assays and stable lines, suggesting a degree of ligand-induced temporal control over expression. However, the authors could use larger sample sizes to further strengthen the robustness of these findings. Nevertheless, his work shows the potential of using agrochemicals in more complex inducible systems, and the authors’ choice to develop control systems using MF as an inducer is an important step towards making plant biology more translational.

Major points 

1. In Figure 3C, the “negative control threshold,” used as a reference to evaluate MF-induced fluorescence in EcR-Cre and EcR-Cre-HIV1 REV stable lines is set by a single unactivated reporter line. Because this threshold value determines which leaf punches show fluorescence above background levels, the authors could consider increasing the number of independent lines, perhaps to n=3. This could capture line to line variability in plants. While the fluorescence readings from the unactivated reporter line do cluster neatly at a low normalized fluorescence value, increasing the sample size would increase the credibility of this threshold.

2. Several experiments rely on selected best performing lines. For example, in Figure 3, the authors select the best performing line of the EcR-Cre and EcR-Cre-HIV1 REV for data display. When investigating silencing by methylation the authors again select single lines of EcR-Cre and EcR-Cre-HIV1 REV. In Figures 5F and 5G, one line with PPP gRNAs is used to calculate the fold change in luminescence compared to plants without PPP gRNAs. Figure 6 references a single DELLA activation line that showed the largest reduction in growth. Overall, the authors could strengthen their data by characterizing multiple lines in each experiment, which would show whether results are robust across independent insertions and help readers understand the range of potential outputs from the systems they develop. They could also consider stating the number of screened plants before selecting the top performers to give insight into how frequently significant activation occurs. Screening large numbers of seedlings could be an important bottleneck for others considering using these systems.

Minor points 

1. The captions of Figures 1 and 2 state, “Across all plots, every dot of the same color corresponds to an independent biological replicate.” The authors could clarify what is meant by biological replicate. For example, in Figure 1E, is each dot a separate leaf infiltrated with the same construct? How many plants were infiltrated, and were they grown on different days or in the same batch? This information would help readers evaluate how replicable the experiment is and what could cause the variability in the data. Similarly, the captions in Figure 5 and 6 refer to biological replicates in stable lines, and additional details would clarify how these replicates are defined.

2. Figure 3E shows representative microscopy images of a root tip from an EcR-Cre HIV1 REV Arabidopsis stable line, untreated and treated with MF. mScarlet expression is higher in the treated root, but the expression pattern is uneven, which is not further investigated by the authors. An explanation would help readers know whether this pattern was expected. The authors could consider visualizing EcR-Cre HIV1 REV by tagging it with a different fluorophore. Overlapping expression of the Cre fusion with mScarlet could point to the T DNA insertion site influencing where the MF activates expression.

3. Bull et al. explore whether silencing due to hypermethylation contributes to lack of normalized reporter fluorescence in Cre recombinase-based stable lines after repeated induction with MF. Supplementary Figure S7 shows partial restoration of EcR-Cre and EcR-Cre-HIV1 REV reporter expression when plants are treated with azaC prior to MF. The authors could consider adding data showing how MF fails to induce fluorescence over time as a direct comparison to recovery with azaC. The statement that “repeated soil-based induction assays using the same Cre recombinase-based stable lines failed to produce significant increases in normalized reporter fluorescence” could also be clarified. For instance, including the generation in which silencing occurs would help readers understand whether methylation builds up over generations as well as how quickly silencing becomes a problem.

4. Figure 4D shows normalized reporter fluorescence in Arabidopsis stable lines across two non-zero MF concentrations. Fluorescence increases as MF increases from 0.5 to 2.2 mM, which the authors state demonstrates the dose-dependent responsiveness of the Cas9-EcR control system. However, only 2.2 mM MF leads to a significant increase in fluorescence, and two non-zero concentrations are typically not enough to demonstrate a dose response. To solidify this claim, the authors could test at least three non-zero concentrations, or ideally more. A curve like this could also be used to identify a saturation point, which would allow the authors to make statements about the maximum output of the Cas9-EcR system.

5. To further support the data in Figure 5, Bull et al. show RT-qPCR data in Supplementary Figure S11 to confirm gene expression changes during the redirection of carbon flux from phenylpropanoid biosynthesis to the FBP. The authors report that “MF treatment resulted in a general decrease in absolute expression levels of C3H, consistent with the previously observed reduction in reporter fluorescence under induced conditions.” This is notable because MF treatment was expected to activate C3H transcription in Cas9-EcR lines transformed with gRNAs and have no effect on lines lacking them. This section would be strengthened if the authors offered a hypothesis for the C3H response. Additionally, seeing as the authors also targeted C4H with gRNAs in this experiment, they could consider showing whether C4H no guide controls are affected by MF treatment in the same manner. Discussing both C3H and C4H expression in no guide controls would add clarity to how much MF dysregulated gene expression in plants, which is an important caveat for its use as an inducer.

Use of AI 

The author declares that they did not use generative AI to come up with new ideas for their review. AI was used during editing to improve grammar and clarity.

I am one of 2 participants from the same research lab that reviewed this preprint as part of the Transparent and Accountable Peer Review training program. This report was written with the support of my research mentor. 

Competing interests

The author declares that they have no competing interests.

Use of Artificial Intelligence (AI)

The author declares that they did not use generative AI to come up with new ideas for their review.