Kinetic properties of optogenetic site-specific DNA recombination by LiCre- loxP
This article has been Reviewed by the following groups
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
- Evaluated articles (Review Commons)
Abstract
Advances in optogenetics now allow to specifically modify the DNA of live cells with light. However, using these technologies successfully requires to know their properties in terms of sensitivity, efficiency, kinetics and mechanism. We previously developed an optogenetic tool made of a single chimeric protein called LiCre that enables the induction of specific changes in the genome with blue light via DNA recombination between loxP sites [1]. Here, we used in vitro and in vivo experiments combined with kinetic modeling to provide a deeper characterization of the photo-activated LiCre- loxP recombination reaction. We find that LiCre binds DNA with high affinity in absence of light stimulus and that this binding is cooperative although not as much as for the Cre recombinase from which LiCre was derived. In yeast, addition of riboflavin to the culture medium had no effect on LiCre’s efficiency, even when cells over-expressed riboflavin kinase, suggesting that abundance of the flavin mono-nucleotide co-factor is not limiting for the reaction. However, LiCre’s efficiency in yeast gradually increased when raising temperature from 20°C to 37°C. The recombination kinetics observed in live cells are best explained by a model where photo-activation of two or more DNA-bound LiCre units (happening in seconds) can produce (in several minutes) a functional recombination synapse. This model was able to capture the effect of a point mutation altering LiCre’s light cycle. This deeper understanding of the LiCre- loxP system provides additional knowledge for designing experiments where specific genetic changes are induced in live cells with light.
AUTHOR SUMMARY
Using a technology called optogenetics, scientists are now able to change the DNA sequence of live cells by illuminating them with light. In theory, they can trigger in genetically-engineered organisms a mutation of interest in specific cells at a specific time. This practice, however, is not common because optogenetics relies on light-controlled enzymes that are recent and not well characterized. We previously developed one of these enzymes, called LiCre , which recognizes a specific piece of DNA. Following illumination with blue light, LiCre can switch on or off whatever gene is close by. Here, we combined experiments and computational modeling to better understand how, and how fast, LiCre works. We find that, although it is inactive in the dark, LiCre does not need photo-activation to bind to its DNA partner. We estimated the speed at which LiCre gets deactivated in the dark and the speed at which active LiCre molecules modify DNA. We also showed the effect of temperature and illumination dynamics on LiCre ’s efficiency. These results will help design strategies where LiCre can be used to conduct genetic studies at high spatio-temporal resolution, or implement it in industrial and biomedical applications.
Article activity feed
-
Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.
Learn more at Review Commons
Reply to the reviewers
The authors do not wish to provide a response at this time.
-
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #3
Evidence, reproducibility and clarity
Summary:
The manuscript by Dufour et al. is a follow-up on the groups' previous publication that introduced the photo-inducible Cre recombinase, LiCre. In the present work, the authors further characterize the properties and kinetics of their optogenetic switch. Initially, the authors show that light affects only LiCre-mediated recombination itself and not DNA binding. Following these observations, they measure and mathematically model LiCre kinetics demonstrating high efficiency in vivo and a surprising temperature sensitivity. Finally, Dufour et al. evaluate several mutations that affect the LOV photo-cycle and provide …
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #3
Evidence, reproducibility and clarity
Summary:
The manuscript by Dufour et al. is a follow-up on the groups' previous publication that introduced the photo-inducible Cre recombinase, LiCre. In the present work, the authors further characterize the properties and kinetics of their optogenetic switch. Initially, the authors show that light affects only LiCre-mediated recombination itself and not DNA binding. Following these observations, they measure and mathematically model LiCre kinetics demonstrating high efficiency in vivo and a surprising temperature sensitivity. Finally, Dufour et al. evaluate several mutations that affect the LOV photo-cycle and provide recommendation for LiCre applications. The study thoroughly investigates various aspects of the function of LiCre, confirming some previously known characteristics (i.e. temperature-dependence of Cre activity and functionality of LOV-based optogenetic tools in yeast without co-factor supplementation), while providing new LiCre-specific insights (kinetics, light-independent DNA binding). Please note that the reviewer is no expert in mathematical modeling and cannot fully judge the methodological details of the models. While I have some concerns as listed below, I believe study should be well-suited for publication after a revision.
Major comments:
- After completing the initial experiment, the authors discovered that their plasmids carry different numbers of V5 epitopes. I am wondering whether this was due to a recombination event happening during the experiment or whether the constructs were not sequence verified prior to use? In any case, an additional ChIP experiment using Cre and LiCre constructs with the identical number of tag-repeats will be necessary. The result, i.e. the strong reduction of DNA-binding of LiCre (which is close to the negative control), is quite remarkable given that LiCre is still considerably active and high DNA affinities were observed in SPR experiments. In light of these counterindications, identical experiment conditions for test and reference group become even more important.
- The conclusion that DNA-binding of LiCre is completely light-independent is not entirely convincing to me. The differences between the light and dark conditions in Fig. 2d are indeed small, but the values for LiCre are almost on par with the vector control and therefore hard to interpret. Based on this experiment alone, one could even be inclined to argue that LiCre does not bind DNA at all (which is of course falsified by the later experiments), showing that the resolution of the corresponding dataset is too low to draw final conclusions. Light-independent DNA binding should either be confirmed by a more sensitive method or the conclusion statements on this matter should be revised accordingly.
- If I understand the explanations correctly, replicates and plotted data points refer to multiple samples (different colonies), that were handled in a single experiment, i.e. by one researcher at the same time/same day. As already mentioned by the authors in the main text, this workflow explains the considerable differences between some of the results in the present manuscript and an identical experiment in a previous publication by the same authors. Providing truly independent experiments (performed on different days) that are therefore independent towards variables such as the fluctuation in incubation temperature (which was the issue in the described experiments) will be crucial, at least for the key datasets.
Minor comments:
- At the end of the Introduction, the authors mention that the interaction of the Cre heptamers was weakened via point mutations in LiCre. A short sentence about the engineering rationale behind this weakened interaction would help readers, who are not familiar with the author's prior work.
- Fig. 2a-b depicts images relating to the purification procedure. These could be moved to the supplements as they don't provide any insight apart from the fact that the proteins were successfully purified.
- The kinetic characterization was only performed for LiCre. Especially for scientists, who have worked with wildtype Cre before, a side-by-side comparison with wt Cre would be valuable to judge the loss in reaction speed that has to be expected when switching from Cre to LiCre.
- The difference between the ChIP results and the SPR results is striking but not mentioned in the discussion section. Also, the statement: "Finally, our results have practical implications on experimental protocols employing LiCre. First, given its high affinity for loxP (Fig. 5b), over-expressing LiCre at high levels will probably not increase its efficiency." (line 502) refers only to the affinity but seems to ignore the low DNA-occupancy of LiCre observed in Fig. 2d. Adapting the discussion section accordingly would improve the manuscript.
Significance
General assessment and advance:
The present study provides a large set of experiments and analyses characterizing the optogenetic LiCre recombinase. In general, the study is well conceived and executed. Although some of my concerns listed above affect key aspects of the study, they should be straightforward to address. The manuscript is a follow-up study providing a more detailed characterization of an optogenetic tool previously developed by the same authors. Its novelty is therefore somewhat limited. While the study provides a rich body of additional data, many of the findings merely confirmed aspects that were to be expected based on the two proteins LiCre is built of (temperature-dependent activity of Cre, optogenetics in yeast w/o the need of co-factor supplementation, weaker DNA-affinity of the Cre fusion protein as compared to wildtype Cre). New insights are provided by the facts that (i) light only controls recombination but not DNA binding and (ii) light activation of only some protomers within the LiCre heptamer is likely to be sufficient to activate recombination. The former aspect is, however, not entirely evident from the results as described above.
Audience:
The study will be of interest for researchers focusing on inducible DNA recombination and especially relevant to those who plan to work with LiCre and can now rely on a more detailed and extended characterization compared to the original LiCre publication.
-
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #2
Evidence, reproducibility and clarity
Summary: This manuscript presents a detailed kinetic and mechanistic characterization of the optogenetic recombinase LiCre, which enables site-specific DNA recombination upon blue-light stimulation. The authors combine in vitro surface plasmon resonance assays, yeast-based recombination assays, and mathematical modeling to dissect the DNA-binding properties, activation dynamics, and recombination efficiency of LiCre. They demonstrate that LiCre binds DNA even in the absence of light, albeit with reduced cooperativity compared to Cre recombinase. Through kinetic modeling, they propose that activation of only two LiCre units may …
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #2
Evidence, reproducibility and clarity
Summary: This manuscript presents a detailed kinetic and mechanistic characterization of the optogenetic recombinase LiCre, which enables site-specific DNA recombination upon blue-light stimulation. The authors combine in vitro surface plasmon resonance assays, yeast-based recombination assays, and mathematical modeling to dissect the DNA-binding properties, activation dynamics, and recombination efficiency of LiCre. They demonstrate that LiCre binds DNA even in the absence of light, albeit with reduced cooperativity compared to Cre recombinase. Through kinetic modeling, they propose that activation of only two LiCre units may suffice for recombination. The study also evaluates the impact of point mutations in the LOV domain on LiCre's photocycle. The experimental methods are described in detail. Statistical analyses are appropriate and clearly reported.
Major Comments:
- In Figure 1, control experiments with no loxP sequences (i.e. original strain) should be performed to demonstrate specific binding of Cre/LiCre to loxP sequence.
- In Figure 2, the SPR experiments are robust and informative. However, the lack of measurement of DNA binding of light-activated LiCre is a notable gap, which will help understand whether the cooperativity of LiCre can be modulated by light. If it is difficult due to experimental conditions, there is lit-mimetic mutant of LOV2 (https://www.nature.com/articles/nmeth.3926).
Significance
General Assessment: This is a rigorous study that combines experimental and computational approaches to advance our understanding of LiCre-based optogenetic genome engineering. The strongest aspects are the integration of SPR data with kinetic modeling and the practical insights into LiCre's performance under various conditions. However, the other limitation is the lack of direct validation of some model predictions.
Advance: To the best of my knowledge, this is the first study to quantitatively model the activation dynamics of LiCre. The work extends previous findings on LiCre and provides new mechanistic and practical insights.
Audience: This study will be of interest to specialized audiences, particularly those developing or applying the LiCre system.
Reviewers' Field of Expertise: Protein engineering, Genome editing, Optogenetics, Cell Biology.
Limitations of Expertise: I do not have deep expertise in mathematical modeling.
-
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #1
Evidence, reproducibility and clarity
Dufour et al describe characterization of the light-activated recombinase LiCre. This work combines the yeast reporter assay, surface plasmon resonance (SPR) and kinetic modeling to provide a comprehensive study of how LiCre functions both in vivo in yeast and in vitro. The authors show that LiCre binds to loxP sites in the dark with high affinity, but reduced cooperativity compared to wild-type Cre, and that recombination efficiency is affected by temperature and illumination regime. Importantly, the authors establish a kinetic model that not only explains these observations but also predicts the altered behavior of a mutant …
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #1
Evidence, reproducibility and clarity
Dufour et al describe characterization of the light-activated recombinase LiCre. This work combines the yeast reporter assay, surface plasmon resonance (SPR) and kinetic modeling to provide a comprehensive study of how LiCre functions both in vivo in yeast and in vitro. The authors show that LiCre binds to loxP sites in the dark with high affinity, but reduced cooperativity compared to wild-type Cre, and that recombination efficiency is affected by temperature and illumination regime. Importantly, the authors establish a kinetic model that not only explains these observations but also predicts the altered behavior of a mutant (T418S), which was experimentally validated. It would be valuable to highlight what other predictions the model could make, even if for future work. Overall, this work combines quantitative experiments and modeling to provide new insights into the biochemical and kinetic properties of LiCre.
Specific comments:
Line 110-115: Although described in the Methods section, a brief statement of dark and light treatment conditions would help readers better follow the experiments. Likewise, listing the three unrelated positions would improve the clarity.
Line 185: Is there a typo?
Line 216: Have the authors considered performing surface plasmon resonance (SPR) to confirm the binding affinity of LiCre-V5 DNA?
Line 233-234: To determine whether the observed difference in recombination efficiency is due to the genomic context of the reporter loci or due to the measurement accuracy of GFP and RFP signals, have the authors considered swapping the positions of GFP and RFP?
Line 236: The sentence "Importantly, we never observed recombination in the entire cell population" is ambiguous. I believe it means recombination was never observed in 100% of the cells. Please rephrase it.
Line 245-249: The hypothesis of plasmid loss based on plating samples on selective and non-selective media without illumination assumes that loss of growth on selective media is only due to plasmid loss, without considering other factors like burden or toxicity. Moreover, the broad range of 10-30% makes it difficult to justify that the ~15% recombination-negative fraction falls within expected variation. The conclusion that LiCre-mediated recombination efficiency is close to 100% after prolonged photoactivation (Line 249, 301-303) is not fully convincing unless more evidence is provided.
Line 275-276: The authors suspect that the decrease in recombination efficiency at very high light intensity is possibly attributed to phototoxicity. Could photobleaching also contribute to this effect? A viability assay would help to validate the phototoxicity explanation.
Line 345-346: While the model with x=2 provides a slightly better fit comparing to the others, the possibility of x=4 cannot be excluded. The inference that "photo-activation of at least two LiCre protomers enables recombination" is not sufficiently proven.
Figure 1e: Please clarify whether the Western blots shown represent biological replicates.
Figure 4: Please include the error bars. Panel a - The authors integrated GFP and mCherry reporters at two different loci to avoid positional bias. Why then is only mCherry used as the ON readout in most experiments, rather than analyzing both reporters in parallel? Please clarify. For panel 4h and line 272, the statement that maximal activation was reached at 12 mW/cm² should be rephrased more cautiously, as no intermediate intensities between 12 and 35.6 mW/cm² were tested.
Significance
This study provides a quantitative experimental and predictive analysis of the light-activated recombinase LiCre, offering new insights into its binding, activation and recombination properties. The predictive validation of the mutant is a strength of this work. While the modeling part is an innovative aspect, more clarification is needed, especially regarding the conclusion that photo-activation of at least two LiCre protomers enables recombination. More mechanistic investigations are needed to support the conclusions. The work will be of interest to researchers in optogenetics, genome engineering, and DNA-protein interactions. My expertise is in yeast genome engineering and applications of Cre-mediated recombination system. Modeling is outside my primary area of expertise.
-
