Rapid Optimization of a Light-Inducible System to Control Mammalian Gene Expression

Synthetic biology tools such as inducible gene expression systems have provided significant contributions to biological research. Their ability to regulate gene expression in a tunable, reversible, and precise manner can improve control of protein production^1,2,3,4,5,6, cell morphology^{7,8,9,10,11,12}, metabolic pathways^{13,14,15,16,17,18}, and other targets for biomanufacturing and therapeutic applications. Chemical inducible systems can have residual^19,20 or off-target effects. Chemical additives can also be very expensive and difficult to scale up industrially due to additional downstream purification processes^21,22,23. Light-inducible gene expression systems can offer a scalable and precise method for biomanufacturing practices^24,25,26,27.

Many light-inducible gene expression systems have been developed as useful synthetic biology tools in a variety of applications^{28,29,30,31,32,33,34,35,36} and wavelengths of blue^35,37,38, red/far-red^12,39,40, or green⁴¹ light. In the case of CRISPR-based optogenetic gene regulation, modifying the amount of individual guide RNAs (gRNAs) delivered can affect the mRNA expression of endogenous genes⁴², and will need to be optimized for different cell lines. Transactivation proteins such as VP64^37,42,43,44, VP16^{39,40,44,45,46}, p65⁴⁴, or fusions of them⁴⁷ can also be used, increasing the modularity of optogenetic systems. In order to investigate complex and intertwined biological pathways^{12,40,48,49,50,51,52,53}, different combinations of these known components can be tested. Additionally, different cell lines can show differences in induction with the same system⁵². Different induction levels can lead to insufficient gene expression or sensitivity in the precise control of gene expression, requiring rigorous testing to find an optimal optogenetic system in novel or more biologically relevant cell lines. With a growing pace and widening applicability of optogenetic gene regulation, it is imperative to rapidly characterize these different systems.

Current methods in characterizing optogenetic tools either use stable or transient expression of genes^12,54. Generating stably expressing cell lines and optimizing system dynamics for maximum expression can be time-consuming and therefore costly. Transient expression enables quick and valuable insights, especially when testing multicomponent optogenetic systems. Here, we used the blue light activated CRISPR-dCas9 effector (LACE)³⁷ system, composed of cryptochrome 2 (CRY2) and cryptochrome-interacting basic-helix-loop-helix N-terminal (CIBN). It has been used to control gene expression in mammalian cells such as HEK293T (human embryonic kidney cells)^37,38, CHO-DG44 (Chinese hamster ovary cells)⁵⁵, and C2C12 (mouse myoblast cells)³⁸. Its modular nature is ideal for quickly characterizing system performance through transient expression and high-throughput methods. For example, multicomponent systems like LACE may need optimization of mass ratios to improve induction, a step not typically employed in optogenetic system characterization.

This protocol details the rapid optimization and characterization of two plasmid LACE (2pLACE)³⁸ system. In this setup, 2pLACE controls expression of a fluorescent reporter, eGFP (enhanced green fluorescent protein), in HEK293T cells. Activation is performed in black glass-bottom 96-well plates using an OptoPlate-96, a 3D printed LED array designed and constructed by Lukasz Bugaj and colleagues^56,57,58,59. This protocol specifically optimizes maximal expression with different mass ratios of the two-plasmid system. It also includes user-friendly code to control different light intensities to investigate the tunability of the system and its full dynamic range. Flow cytometry is used for data collection and analysis. Protocols for generating plasmid constructs and 3D printing materials for the LED array can be found in previous publications^54,56. These methods highlight a quick, high-throughput pipeline for characterizing light-inducible gene expression systems.

1. Plating HEK293T cells in a 96-well format

1. In a biosafety cabinet, collect a bleach waste container, serological pipettes, pipette-aid, Dulbecco's Phosphate Buffered Saline (DPBS), Dulbecco-modified Eagle's Medium (DMEM) with 10% fetal bovine serum (FBS), and a T-75 cell culture flask with confluent HEK293T cells. Prewarm the media to 37 °C.
2. In a T-75 flask, check that the confluency is about 80%-100% confluent through an inverted microscope.
3. In the biosafety cabinet, use a serological pipette to aspirate the spent media from the cell culture flask. Avoid touching the surface or plastic of the flask. Dispose of the spent media and aspirator in the waste container.
4. Gently wash the cells with about 10 mL of DPBS by aspirating it into a corner of the flask and gently swirling it around the surface. Aspirate the DPBS from the flask and dispose of the wash and the aspirator in the waste container.
5. Add 1.5 mL of 0.05% Trypsin-EDTA to cover the surface of the flask and incubate at room temperature for 1 min. Gently tap the flask to loosen the cells from the surface.
6. Add 8.5 mL of fresh DMEM into the flask. Resuspend and mix cells with a serological pipette until all aggregates have been broken up by aspirating up and down.
7. Collect 9 mL of cell suspension into a 15 mL conical tube.
   1. Add 9 mL of fresh DMEM into the flask and place it into an incubator at 37 °C and 5% CO₂
8. Using a hemocytometer, mix 10 µL of suspended cells with 10 µL of Trypan blue dye at a 1:1 ratio to calculate the concentration of cells. Use this value to prepare enough cells to seed on the plate.
9. Seed ~35,000 cells in 100 µL for each well of a high-performance #1.5 black 96-well glass bottom plate and place in an incubator at 37 °C and 5% CO₂ for 24 h.

2. 2pLACE transfection optimization

1. For one well and 10% excess, aliquot 11 µL of warm serum-free DMEM (SFM) into a 1.5 mL microcentrifuge tube.
2. Using a CRY2-eGFP: CIBN-gRNA plasmid mass ratios equal to 1:9, 2:8, 3:7, 4:6, 5:5, 6:4, 7: 3, 8:2, and 9:1, aliquot 110 ng for each well of total DNA to the aliquots of SFM (Table 1).
3. As a positive control, aliquot the same mass of CMV-eGFP plasmid to different aliquots of SFM.
4. For each well, aliquot 11 µL of SFM for transfection reagent dilution.
5. Prepare the transfection reagent at a 1:3 mass (µg) to volume (µL) ratio of DNA and transfection reagent.
6. Add diluted transfection reagent to the diluted DNA and incubate for 12 min at room temperature to create the transfection complexes.
7. Add ~20 µL of the transfection complex to each well. Ensure that it does not touch the walls of the well.
8. Wrap the plate in aluminum foil and place it in an incubator at 37 °C and 5% CO₂ for 24 h.

3. 2pLACE activation

1. Modify the microcontroller input code to illuminate the desired wells using these settings (Supplementary Coding File 1, Lines 147 - 158).
   1. LED Intensity: 9.27 mW/cm² (Supplementary Coding File 1, Lines 200 - 215).
   2. Pulse Length: 1 s (Supplementary Coding File 1, Lines 280 - 290).
   3. Pulse Frequency: 0.067 Hz (every 15 s) (Supplementary Coding File 1, Lines 264 - 278).
      ​NOTE: Wavelength settings are determined by the types of LEDs installed.
2. Spray the 3D printed lid of the OptoPlate with 70% ethanol and let it dry in a biosafety cabinet.
3. Turn on a darkroom red light lamp and place the 96-well plate into a biosafety cabinet, and replace the lid with the dried 3D printed lid.
   NOTE: One can optionally view the CMV-eGFP cells using fluorescence microscopy to estimate transfection efficiency.
4. Take the 96-well plate and place it onto the LED array to construct the LED array apparatus. Ensure that the plate snaps into place.
5. Connect the microcontroller, LED, and fan ports to a power source.
6. Carefully spray wires down with 70% ethanol and wipe dry.
7. Place the LED array apparatus into an incubator at 37 °C and 5% CO₂ for 24 h. Ensure that the LEDs turn on as intended and that the wires are not taut when the incubator is sealed.

4. Flow cytometry preparation

1. While in a red-light environment, take out the OptoPlate and place the plate into a biosafety cabinet
   NOTE: One can optionally view the cells by fluorescence microscopy to visually confirm light induction.
2. Carefully aspirate all media from each well.
3. Detach the cells using 30 µL of 0.05% Trypsin-EDTA and incubate for 1 min at room temperature.
4. Mix each well with 100 µL cold FACS buffer (1% FBS in PBS) until single-celled.
5. Transfer each sample into a V-bottom 96-well plate.
6. Centrifuge the V-bottom 96-well plate at 164 x g for 10 min at room temperature.
7. Aspirate 80 µL of supernatant without disturbing the pellet.
8. Add 200 µL of cold FACS buffer to resuspend the pellet.
9. Wrap the V-bottom 96-well plate in aluminum foil for light insulation.
10. Place the plate into a styrofoam box with ice for transportation.

5. Flow cytometry gating and data collection

1. Turn on the flow cytometer using the switch in the back, and open the respective flow cytometry software on the computer.
2. Run the system startup program and load 2 mL of DI water in the sample loader (Supplementary Figure 1A).
3. Run quality control (QC) using the provided QC beads (Supplementary Figure 1B).
4. Ensure the flow cytometer is in plate sampling mode (Supplementary Figure 1C).
5. Set up and specify sample wells (Supplementary Figure 1D).
6. Create the following scatter plots and tables (Supplementary Figure 2A).
   1. Side Scatter (SSC-A) vs Forward Scatter (FSC-A).
   2. SSC-H vs SSC-A.
   3. SSC-A vs FITC-A.
   4. FITC-A statistics table (Supplementary Figure 2B).
7. Snap the V-bottom plate into the plate loader of the cytometer.
8. Select an untransfected well and click on Initialize, and then click on Run.Ensure volume sample rate is 10 µL/min (Supplementary Figure 2C).
9. Adjust SSC and FITC voltages to center the population of interest on the SSC-A vs FSC-A plot (Supplementary Figure 2C).
   1. Create a polygon to gate the healthy cell population of interest (Supplementary Figure 2D).
   2. Create a polygon to gate for doublet discrimination on the SSC-H vs SSC-A plot.
   3. Adjust FITC voltage to place untransfected cells towards the left of the SSC-A vs FITC-A plot (Supplementary Figure 2D).
10. Repeat for the remaining untransfected wells and record Mean FITC-A data.
11. Select a CMV-eGFP transfected well and click on Run.
12. Adjust FITC voltage to capture both autofluorescing and fluorescing cells in the SSC-A vs FITC-A plot.
    1. Create a polygon to gate the fluorescing cells against the non-fluorescing cells while referencing the initial gate in untransfected cells (Supplementary Figure 2E).
13. Auto-record samples at 60 µL/min until 200 s have been reached and/or events have reached 10,000 (Supplementary Figure 2F).
14. Export Mean FITC as a CSV file and analyze the data.
15. Clean the flow cytometer through the Daily Clean option (Supplementary Figure 2G).

6. LED intensity optimization

1. Edit the microcontroller script to test desired LED intensities (Supplementary Coding File 1, Lines 200-215).
   NOTE: Equivalent LED intensities with respect to the values in the microcontroller script are reported elsewhere³⁸. Self-calibration may be necessary when using different LEDs.
2. Ensure the script contains the following values in (Supplementary Coding File 1, Lines 200-215) as designed in Table 2, and upload the code to the microcontroller.
3. Repeat steps 1-5.

Adopting workflow elements from previous literature37,38,55, we added high-throughput simultaneous illumination from the OptoPlate-96 and demonstrated a pipeline to rapidly optimize a light-inducible gene expression system in mammalian cells (Figure 1).

For inducible gene expression systems, high dynamic ranges are a critical performance marker, in addition to absolute on and off (leaky) expression. With fluorescence imaging of transfected cells, the difference in eGFP expression between the light-activated cells and the cells kept in the dark can be qualitatively observed (Figure 2). A ratio of 1:9 visibly results in less maximal eGFP expression compared to a ratio of 5:5. It can also be observed that there is increased leakiness in the 5:5 ratio. Fluorescence imaging of a constitutively expressing eGFP (CMV-eGFP) transfected cells and untransfected cells are valuable positive and negative controls to contextualize transfection efficiency and the system's induced and leaky expression, respectively. Using fluorescence imaging allows users to quickly see and validate successful transfection and activation of gene expression with blue light, providing a valuable checkpoint before flow cytometry. Flow cytometry can then be used to quantify the mean fluorescence intensity (MFI) and the dynamic range.

After light-activating the cells, they are prepared with FACS buffer and analyzed through flow cytometry. HEK293T cells are about 11-15 µm, resulting in an FSC gain of 500 and an SSC gain of 125 (Supplementary Figure 2C) to center the healthy cell populations. Higher voltages result in analysis of debris rather than healthy cell populations. Our runs consistently have above ~60% of events in the first population gate (P1) (Figure 3A-D). Once the healthy cell populations make up most of the events in the SSC-A vs FSC-A plot, the density of events can also be checked to identify the majority of the population (Supplementary Figure 3A). Then, doublet discrimination can be performed through the SSC-H vs SSC-A plot, which is crucial for reliable and reproducible results60,61. In this system, we generally found more than 95% of events gated in P1 to be singlets (Figure 3A-D). In the SSC-A vs FITC-A plot, untransfected cells will generally be towards the left of the center of the plot and be gated to have a population <0.1% (Figure 3A). Then, eGFP-positive cells will populate the gate that was empty in the untransfected sample, resulting in single-cell analysis measurements of MFI (Figure 3B). Once the negative and positive controls are collected to set proper gates and voltages, samples with 2pLACE can be analyzed using the same gates to reliably exclude autofluorescing cells (Figure 3C,D). Here, the percent population for eGFP-positive cells was ~99% for CMV-eGFP and ~57% for 2pLACE-transfected cells. A final gate can also be used on the PE-A channel to ensure highly fluorescent cells are captured (Supplementary Figure 3B).

Another ratio that was opted to test was 3:7. As a validation step, fluorescence microscopy images were taken before flow cytometry and observed successful transfection for the induction of eGFP expression with 2pLACE, expectedly less than CMV-eGFP (Figure 4A). After collecting flow cytometry data of 2pLACE, blue light-activated gene expression can be compared with leaky gene expression (Figure 4B). Using the settings listed in Step 3.1, we were able to observe a ~3-fold increase in expression following blue light activation, matching the increased fluorescence signal seen qualitatively by microscopy. Additionally, transfection efficiencies can be indirectly seen by measuring the % eGFP-positive population, or percent parent, at ~60% of healthy single cells (Figure 4C).

Implementing this protocol in its entirety can identify optimal mass ratios (Figure 5A) and light intensities (Figure 5B) for maximum dynamic range and tunable expression. Mass ratios lower than 6:4 exhibited larger dynamic range (3.5-4.5), while mass ratios beyond 6:4 showed decreased dynamic range due to increased background and decreased maximal eGFP expression. For both maximal expression and high dynamic range, a ratio of 3:7 is preferred. Ratios below 3:7 showed similar fold inductions but had less overall maximal expression. Using our previously calibrated light intensity values38 (Table 2), one can characterize gene expression level with respect to different light intensities. Light intensities also controlled expression levels of eGFP, where MFI saturates at ~3 mW/cm2. Light intensities beyond this may have varying MFI values, likely due to photobleaching in some samples as seen at ~9 and 11 mW/cm2.

Figure 1: General pipeline of high-throughput experiments testing optogenetic systems. Cells are seeded into a black 96-well plate for 24 h. Then, cells are transfected with a 12 min incubation period of DNA and transfection reagent. 24 h after transfection, the plate is placed onto the OptoPlate-96 for blue light activation. After the desired duration of blue light activation, cells are prepared in a red light environment for flow cytometry. Flow cytometry data is represented as mean fluorescence intensity graphs of eGFP-positive cells in the dark or treated with blue light. Please click here to view a larger version of this figure.

Figure 2: Representative fluorescence microscopy images of 2pLACE in the on and off state in HEK293T cells. Different mass ratios of CRY2-eGFP: CIBN-gRNA were tested for levels of eGFP expression. Cells were exposed to blue light (ON) or kept in the dark (OFF) for 24 h. The scale bar represents 100 µm. Please click here to view a larger version of this figure.

Figure 3: Representative flow cytometry scatter plots of untransfected (UT), CMV-eGFP, and 2pLACE HEK293T cells. Flow cytometry was performed using the plate loader function. (A) HEK293T cells were gated based on forward scatter (FSC-A) and side scatter (SSC-A). The concentration of events on SSC-A vs. FSC-A plots can be visualized using a contour plot (Supplementary Figure 3A) to assist with gating healthy populations. Doublet discrimination was performed on an SSC-H vs. SSC-A plot using a linear gate. The final plot gated fluorescent cells by referencing the untransfected sample. (B) Cells were gated as in (A), showing the fluorescent population and fluorescence intensity in the statistics table (Supplementary Figure 2B). (C,D) 2pLACE-transfected cells were gated separately through the P3 gate. The magenta population represents all eGFP-positive cells (Supplementary Figure 3B). Please click here to view a larger version of this figure.

Figure 4: Qualitative and quantitative analysis of light-induced eGFP expression in HEK293T cells. (A) Fluorescence microscopy images of HEK293T cells transfected with either 2pLACE or CMV-eGFP plasmids. (B) Quantification of mean fluorescence intensity (MFI) of eGFP in cells maintained under blue light or dark conditions. (C) Percentage of eGFP-positive cells measured through flow cytometry gating analysis. Gating resulted in <0.1% of untransfected cells falling within the eGFP-positive threshold. Flow cytometry data represent mean values, and error bars indicate standard deviations from four technical replicates. Please click here to view a larger version of this figure.

Figure 5: Flow cytometry of representative mass ratios of 2pLACE and blue light intensities. (A) Various mass ratios of CRY2-eGFP: CIBN-gRNA were transfected into HEK293T cells, which were then exposed to blue light or kept in the dark using the same settings as reported in step 3 of the protocol. Each condition is the mean of 3 biological replicates. (B) Representative trend of gene expression as a function of blue light intensity. Each condition is the mean of 6 technical replicates. Mean fluorescence intensities are quantified through flow cytometer gating of cells expressing eGFP. All error bars represent standard deviation. For plasmid ratios, MFI statistical significance was calculated using a one-way t-test comparing the light and dark of a plasmid ratio. For light intensities, statistical significance was calculated with one-way ANOVA and Tamhane's T2 post-hoc test compared to the dark (OFF) state. *p < 0.05, **p < 0.01, ***p < 0.001, ****p <0.0001, *****p < 0.00001 and ns is not significant. This figure was adapted from Garimella et al.38. Please click here to view a larger version of this figure.

Table 1: Amount of CRY2-eGFP and CIBN-gRNA plasmids per well using example concentrations. Volume amounts are per well and can be modified based on the number of samples or replicates in an experiment. Please click here to download this Table.

Table 2: Recommended values for blue light intensities. Code is found in Supplementary Coding File 1 (Lines 215-228). Equivalent intensity values are calculated using calibration data previously reported using an optical power meter. Each intensity level has 6 technical replicates. Rows G and H are intended to be for CMV-eGFP and untransfected control wells. Please click here to download this Table.

Supplementary Figure 1: Steps 5.1-5.5 of the protocol. (A) Running the system startup program. (B) Quality control standardization using fluorophore beads. (C) Switching from tube sampling to plate loader mode. (D) Selection of wells to label as sample wells; 35 wells were selected for this example. Please click here to download this File.

Supplementary Figure 2: Steps 5.6-5.14 of the protocol. (A) Setting up light-scattering plots: FSC-A vs. SSC-A for healthy cell population gating, SSC-A vs. SSC-H for doublet discrimination, and SSC-A vs. FITC-A for gating autofluorescent and eGFP-positive cells. (B) Setting up the statistics table to acquire FITC-A data to measure mean fluorescence intensity. (C) Highlighting stopping rules in the Events to Display tab. Ensure a slow sample flow rate, eject and load the plate, initialize the flow cytometer probe, and adjust acquisition voltages. (D) Creating polygons to gate each population as shown. (E) Adding the appropriate number of wells. (F) Clicking on Auto-record once settings and gates are complete. (G) Clicking on Daily Clean after all samples are collected and statistics are exported. Please click here to download this File.

Supplementary Figure 3: Validation of all cells within the captured gate, visualized by green events in P4 (magenta). (A) Contour plot of the healthy population within the eGFP-positive population gate (P3). Cells are mainly concentrated in the red zones, decreasing outwardly. (B) SSC-A vs. PE-A plot for gating P4, providing an additional check to ensure all fluorescing events are accounted for in the FITC-A plot. Please click here to download this File.

Supplementary Coding File 1: Example Arduino code implementing all properties described in the above protocol. This file activates all LED positions using the recommended LED intensity values in Table 2. Please click here to download this File.

This protocol demonstrates the optogenetic control of a multicomponent transient gene expression in mammalian cells, allowing other researchers to rapidly and simultaneously characterize modular optogenetic systems with a fluorescent reporter.

In this case, a two-plasmid light-inducible system was optimized for the expression of eGFP in HEK293T cells. For successful co-transfection, it is essential to optimize seeding densities and mass ratios for the highest transfection efficiency. This is especially crucial when implementing this in stem cell culture, where confluency can induce differentiation^62,63,64 and would lead to inconsistencies in data analysis if the cell culture is not homogenous. In cases where the stemness of the cells are maintained, it may be necessary to seed at lower densities and replace the media before transfection, as media composition can also differentiate cells^{65,66,67,68,69}. Multicomponent light-inducible gene expression systems can have varying performances depending on the mass ratio of each plasmid^38,70. Generally, due to the minimal CMV promoter to express eGFP, increasing the ratio of CRY2-eGFP to CIBN-gRNA can lead to increased leakiness and decreased dynamic range. Alternatively, this protocol can be applied when optimizing single-construct light-inducible systems, such as dSpCas9⁷¹ and iLight⁴⁶, in other mammalian cell lines, which may especially be desirable in hard-to-transfect cells such as C2C12 mouse myoblasts. Common methods for evaluating hard-to-transfect cells involve generating a stable cell line with the system, either virally or non-virally. Viral transduction and transposition-based cell line generation previously resulted in functional optogenetic regulation of gene expression and could be considered when working with primary and stem cells.

While this method demonstrates optogenetic control of a fluorescent transgenic reporter as an example, CRISPR technology allows targeting of various endogenous genes^37,48,49,52, which can be useful for researchers and industry developing gene-based therapeutic solutions or in biomanufacturing¹⁹. If endogenous genes are of interest, immunofluorescence, quantitative real-time PCR, and other gene-specific assays can be employed to characterize and quantify light-inducible gene expression. However, these gene-specific assays can have varying degrees of compatibility with high-throughput analysis depending on the equipment available.

For semi-adherent mammalian cells such as HEK293T cells, the seeding density in a glass-bottom 96-well plate can affect the percent positive eGFP events analyzed in the flow cytometer. It is desired to achieve high %eGFP positive events to ensure sample uniformity and enough cells of interest to measure in flow cytometry. Transfection efficiency can effectively be lower in overconfluent cultures⁷², illustrating the need to optimize seeding densities.

In troubleshooting flow cytometry preparation, ensuring that cells do not get washed away throughout the preparation step and having both negative and positive controls are key to successful data collection. Since HEK293T cells are semi-adherent cells, it is likely that during trypsinization and D-PBS washing, many cells will be lost. To potentially save time, it may be helpful to briefly check each well through a fluorescence microscope between each step, which may cause a significant loss in the number of cells. Additionally, the negative and positive controls used here are an untransfected and a constitutive CMV-eGFP sample, providing a way to measure autofluorescence of the HEK293T cells during flow cytometry and to see if transfection has worked for the experiment. While this protocol uses untransfected samples for autofluorescence, an optional control for background to employ would be transfecting only the CRY2-eGFP vector, assuring that there is no additional background. The CMV-eGFP positive control can provide a quick readout through fluorescence microscopy 24 h post-transfection, as it may be difficult to gauge if the LACE system was transfected successfully before any blue light treatment. If CMV-eGFP does not have the expected transfection efficiencies, it may be necessary to consider a different transfection reagent for the experiment.

When collecting flow cytometry data for the first time, it may be necessary to prepare extra positive and negative control cells. This will help ensure there is enough sample to adjust flow cytometer settings. Preparing extra untransfected cells from the same T-75 flask used to plate cells can also provide a useful comparison for assessing cell health between the untransfected cells that were either seeded in the plate or in the flask. When analyzing the untransfected cells to set FSC and SSC voltage gates, it is crucial that the flow cytometer is collecting at a slow flow rate (i.e., 10 µL/min). A slow flow rate preserves accurate and consistent data collection and ensures enough cells can be analyzed for proper gating. When gating CMV-eGFP positive cells against autofluorescing cells, adjusting the FITC-A voltage may be tedious as eGFP can exceed the log scale, even at the lowest FITC-A voltage. Voltage values between 1 and 10 generally work in gating most of the light-induced eGFP-positive cells in HEK293T cells.

For troubleshooting optimal light intensity conditions, the values in Table 2 may cover all needed values for blue light systems. Blue light systems are inherently very sensitive^12,71,73,74 and do not need high intensity. The max intensity setting at 4095 for the OptoPlate-96 can be used as the most phototoxic and most photobleaching condition. This range of intensities can also provide insights into phototoxic effects of a range of blue light exposure using cell viability stains like propidium iodide^75,76. It is important to note that when working with fluorescent protein expression, photobleaching can occur when exposed to high-intensity blue light and can negatively affect results. If noticeable photobleaching occurs, using lower intensities of blue light can give better results. Previous literature utilizing the LACE systems in HEK293T and CHO-DG44 cells used the LED intensities and time course reported in this protocol^37,55. Different mammalian cell lines may have different thresholds when exposed to the conditions used here, possibly resulting in low viability^54,77,78,79. For other blue light systems, such as LOV2⁸⁰ or EL222⁸¹, the frequency or pulse duration of blue light exposure could also be adjusted based on the kinetics of protein dimerization⁸². If blue light is incompatible with the cells of interest through noticeable cytotoxicity or photobleaching, it may be necessary to switch to a longer wavelength inducible system, such as REDLIP³⁶ or iLight2⁸³.

Proper activation of an optogenetic tool also needs minimal leaky expression. This protocol has the potential to introduce undesired light exposure to the cells. With the fluorescence imager and flow cytometer, blue light is used to validate and measure fluorescence, respectively. However, in the case of 2pLACE, significant activation compared to the dark state occurred at 30 min of blue light activation³⁸, demonstrating that using this tool does not require the perfect darkroom conditions that we employ using a red light lamp. Therefore, by limiting the duration of exposure to light sources by using aluminum foil, leaky expression can be reduced to a minimum. The leaky expression we observe may originate from the design of the CRY2-eGFP plasmid. A previous study reduced background expression by reducing the linker copies in the original iLight construct⁸³, and this may be a method of interest to optimize 2pLACE in the future. Additionally, this method utilizes transient transfection, which can result in variable levels of gene expression across the same population due to differences in copy numbers received. This variability could be reduced by creating a stable cell line using 2pLACE, but this process can be slow to implement. Until further work is performed on reducing leaky or variable expression, transient transfection may not be reliable enough for certain downstream applications, but can serve as a quick method for representative performance of optogenetic regulation of gene expression.

With high-throughput devices enabling an array of different conditions in an efficient manner, optogenetic tools in biotechnology research can accelerate investigations in targeting biological pathways. Transient expression offers a non-viral and rapid method to characterize synthetic biology tools. Further improvements could be to implement computational tools such as Python or RStudio for faster data analysis in high-throughput assays.