Spatiotemporal Subcellular Manipulation of the Microtubule Cytoskeleton in the Living Preimplantation Mouse Embryo using Photostatins

Summary

Typical microtubule inhibitors, used widely in basic and applied research, have far-reaching effects on cells. Recently, photostatins emerged as a class of photoswitchable microtubule inhibitors, capable of instantaneous, reversible, spatiotemporally precise manipulation of microtubules. This step-by-step protocol details the application of photostatins in a 3D live preimplantation mouse embryo.

Abstract

The microtubule cytoskeleton forms the framework of a cell and is fundamental for intracellular transport, cell division, and signal transduction. Traditional pharmacological disruption of the ubiquitous microtubule network using, for instance, nocodazole can have devastating consequences for any cell. Reversibly photoswitchable microtubule inhibitors have the potential to overcome the limitations by enabling drug effects to be implemented in a spatiotemporally-controlled manner. One such family of drugs is the azobenzene-based photostatins (PSTs). These compounds are inactive in dark conditions, and upon illumination with UV light, they bind to the colchicine-binding site of β-tubulin and block microtubule polymerization and dynamic turnover. Here, the application of PSTs in the 3-dimensional (3D) live preimplantation mouse embryo is set out to disrupt the microtubule network on a subcellular level. This protocol provides instructions for the experimental setup, as well as light activation and deactivation parameters for PSTs using live-cell confocal microscopy. This ensures reproducibility and enables others to apply this procedure to their research questions. Innovative photoswitches like PSTs may evolve as powerful tools to advance the understanding of the dynamic intracellular microtubule network and to non-invasively manipulate the cytoskeleton in real-time. Furthermore, PSTs may prove useful in other 3D structures such as organoids, blastoids, or embryos of other species.

Introduction

The microtubule architecture varies widely across different cell types to support diverse functions^1,2. Its dynamic nature of growth and shrinkage allows rapid adaptation to extra- and intracellular cues and to respond to the ever-changing needs of a cell. Hence, it can be considered as the "morphological fingerprint" playing a key role in cellular identity.

Pharmacological targeting of the microtubule cytoskeleton using small molecule inhibitors has led to a plethora of fundamental discoveries in developmental biology, stem cell biology, cancer biology, and neurobiology^3,4,5,6,7. This approach, while indispensable, presents various limitations such as toxicity and off-target effects. For example, one of the most widely used microtubule-targeting agents, nocodazole, is a powerful microtubule-depolymerizing drug⁸. However, small-molecule inhibitors such as nocodazole are active from the time of application and, given the essential nature of the microtubule cytoskeleton to many critical cellular functions, global depolymerization of microtubules can produce off-target effects, which may be unsuitable for many applications. Additionally, nocodazole treatment is irreversible unless samples are washed free of the drug, preventing continuous live imaging and, thus, precise tracking of individual microtubule filaments.

The development of light-activated compounds began with the creation of photouncaged molecules and has heralded a new era in targeting and monitoring the effects of microtubule growth inhibition in a precise and spatiotemporally-controlled manner. One family of reversibly photoswitchable drugs, photostatins (PSTs), were developed by replacing the stilbene component of combretastatin A-4 with azobenzene⁹. PSTs are inactive until illumination with UV light, whereby the inactive trans-configuration converts to the active cis-configuration by reversible isomerization. Cis-PSTs inhibit microtubule polymerization by binding to the colchicine binding site of β-tubulin, blocking its interface with β-tubulin and preventing dimerization required for microtubule growth¹⁰. Among a cohort of PSTs, PST-1P has emerged as a lead compound as it has the highest potency, is fully water-soluble, and shows a rapid onset of bioactivity after illumination.

The most effective trans- to cis-isomerization of PSTs occurs at wavelengths between 360-420 nm, which enables dual options for PST activation. A 405 nm laser line on a typical confocal microscope can be administered for optimal spatial targeting of microtubule growth inhibition. The ability to pinpoint the location and timing of PST activation through 405 nm laser illumination facilitates precise temporal and spatial control, allowing disruption of microtubule dynamics on a subcellular level, within sub-second response times⁹. Alternatively, an affordable LED UV light allows whole organism illumination to induce organism-wide disruption of the microtubule architecture. This may be a cost-effective alternative for researchers for whom the precisely-timed onset of inhibition, rather than spatial targeting, is the goal. Another feature of PSTs is their on-demand inactivation by applying green light of a wavelength in the 510-540 nm range⁹. This enables tracing of microtubule filaments before, during, and after PST-mediated growth inhibition.

PSTs, while still a relatively recent design, have been used in numerous in vitro applications across diverse research fields¹¹, including investigating new mechanisms of cell migration in amoeboids¹², in neurons isolated from the brain of the newborn mouse¹³, and wing epithelium development in Drosophila melanogaster¹⁴. Other light-reactive drugs have proven to be valuable tools in targeted disruption of cellular function. For instance, an analog of blebbistatin, azidoblebbistatin, was used for enhanced myosin inhibition under illumination^15,16. This highlights the potential for new discoveries owing to the ability for spatiotemporally controlled inhibition of cellular function.

Live 3D organisms present superb yet more delicate systems to manipulate microtubule dynamics on a whole-animal, single-cell, or subcellular level under physiological conditions. In particular, the preimplantation mouse embryo offers exceptional insight into the inner workings of the cell as well as intercellular relationships within an organism¹⁷. Temporally and spatially targeted consecutive cycles of activation and deactivation of PSTs contributed to the characterization of the interphase bridge, a post-cytokinetic structure between cells, as a non-centrosomal microtubule-organizing center in the preimplantation mouse embryo¹⁶. A similar experimental setup demonstrated the involvement of growing microtubules in the sealing of the mouse embryo to allow blastocyst formation¹⁸. Furthermore, PSTs were also used in whole zebrafish embryos to investigate neuronal cell migration by inhibiting microtubule growth in a subset of cells in the hindbrain¹⁹.

This protocol describes the experimental setup and use of PST-1P in the preimplantation mouse embryo. The instructions presented here can also guide the application of PSTs for a wide array of objectives such as studying chromosome segregation and cell division, trafficking of intracellular cargo, and cell morphogenesis and migration. Furthermore, such studies will assist the implementation of PSTs in organoid systems, blastoids, and other embryo models such as Caenorhabditis elegans and Xenopus laevis, as well as potentially expand the use of PSTs for in vitro fertilization technologies.

Protocol

Experiments were approved by the Monash Animal Ethics Committee under animal ethics number 19143. Animals were housed in specific pathogen-free animal house conditions at the animal facility (Monash Animal Research Platform) in strict accordance with ethical guidelines.

1. Preimplantation mouse embryo collection

1. Superovulate and mate mice as described previously^16,18, in compliance with the institutional animal ethics guidelines.
   NOTE: The most commonly used strains for live embryo collection are C57BL/6 or FVB/N mice. All data shown here were generated using FVB/N mice.
2. On the morning after mating, flush zygotes from the oviduct using M2 medium as described²⁰, or Human Tubal Fluid (HTF) medium. Using a mouth pipette apparatus as described^21,22, transfer zygotes to fresh Potassium Simplex Optimised Medium (KSOM) droplets, prewarmed to 37 °C and equilibrated to 5% CO₂, in a 35 mm culture dish overlaid with a sufficient volume of mineral oil to ensure media coverage.
3. Microinject zygotes as described²⁰ with cRNA encoding for a red fluorescently-tagged microtubule plus end marker. Here, cRNA for the End Binding protein 3 (EB3)-dTomato was used at 30 ng/µL concentration after preparing and purifying as described^16,18 and diluting in microinjection buffer.
   NOTE: It is recommended to prepare cRNA ahead of time and store it at -20 °C until required.
4. Culture embryos in dark at 37 °C and 5% CO₂ until embryos have reached the desired developmental stage for PST-1P treatment.
   NOTE: For a comprehensive resource of the culture times required for different embryonic stages see²³. For 16-cell stage embryos used here, culture to embryonic day 3 (E3) post-fertilization.

2. Drug and imaging dish preparation

NOTE: For Steps 2.1-2.10, work exclusively in the dark or red-light conditions to avoid unintended PST-1P activation. Aluminum foil or dark covers should be used for all tubes and dishes containing PSTs.

1. Prepare a stock concentration of 50 mM PST-1P in ultrapure water.
   NOTE: The molecular weight of PST-1P is 440 g/mol. The stock solution is stable at -20 °C for up to 1 year. PST-1P is soluble in water or aqueous buffer but does not readily dissolve in DMSO.
2. From step 2.1, prepare an intermediate working concentration of 800 µM PST-1P in ultrapure water.
3. Dilute PST-1P to a final concentration of 40 µM in fresh KSOM. As a typical experiment requires approximately 20 µL of PST-1P-treated KSOM, dilute 1 µL of 800 µM PST-1P in 19 µL of KSOM for a final volume of 20 µL to ensure sufficient medium is prepared in advance, using only red light for visibility.
   NOTE: Both the concentration and the activation status of a PST-1P dilution can be checked by taking a UV-Vis absorbance spectrum using a spectrophotometer which should be done during assay establishment. The absorbance at 380 nm (A₃₈₀) of a fully-trans 40 µM dilution in a 1 cm cuvette should be approximately 0.8. Both cis- and trans-forms have the same absorbance at 455 nm (A₄₅₅). When the ratio of the A₃₈₀ to the A₄₅₅ is approximately 9:1, the dilution is inactive (fully trans). When the ratio A₃₈₀:A₄₅₅ is 1:2, the dilution is fully activated (fully cis). Intermediate ratios reflect intermediate states of activation.
4. To prepare the chamber slide for live imaging, pipette 10 µL of PST-1P-treated KSOM into the center of one well to form a hemispherical droplet (Figure 1A).
5. Gently add sufficient mineral oil to cover the droplet, ensuring that it does not disperse the media. This will ensure that the droplet does not evaporate.
6. Pre-warm and CO₂-equilibrate the chamber slide dish in an incubator at 37 °C and 5% CO₂ for a minimum of 3 h or at most, overnight.
7. At the end of the equilibration period in step 2.6, prepare a 35 mm culture dish with a 10 µL droplet of PST-1P-treated, prewarmed, and equilibrated KSOM as a wash step. Do not overlay with oil.
8. Transfer embryos by mouth pipetting into the PST-1P-treated KSOM droplet from step 2.7.
   NOTE: Steps 2.7 and 2.8 are recommended for mouth pipetting of embryos but are optional.
9. Immediately transfer embryos by mouth pipetting into the center of the PST-1P-treated KSOM droplet in the imaging chamber slide prepared in steps 2.4-2.6 (Figure 1A).
10. Incubate embryos at 37 °C and 5% CO₂ in PST-1P-treated KSOM in the imaging chamber slide for at least 1 h before imaging. If possible, mount the chamber slide on the microscope inside an environmental chamber at 37 °C and 5% CO_2, and in complete darkness to ensure all PST-1Ps are in the inactive trans-configuration and that embryos can sink to the bottom of the dish.

3. Live imaging and PST-1P photoactivation

NOTE: Steps 3.1-3.13 are performed on a laser scanning confocal microscope fitted with avalanche photodiode detectors (APDs) and a dark environmental chamber. These instructions refer specifically to the imaging setup using the acquisition software described in the Table of Materials; however, they can also be applied to other confocal microscopy systems.

1. Prepare a 63x/1.2 NA water oil immersion objective with the prescribed immersion medium.
2. Using a red light torch to guide positioning, advance the objective to contact the immersion medium. At this stage, avoid using white or brightfield light to find the embryos as this could precociously activate PST-1P.
3. Using the eyepiece and while under red light illumination, locate the edge of the droplet of medium and position the objective directly over this location. This can assist the user to establish orientation and find the focal plane.
4. Next, through the eyepiece or on software-enabled live-mode scanning, use a red wavelength filter or 561 nm laser to locate the embryos within the droplet.
5. Use stage controllers and live-scanning mode to set the start- and end-points for acquiring a z-stack of the whole embryo.
6. Adjust the laser power settings (typically, with highly sensitive detectors such as the APDs, a 561 nm laser power of less than 5% is sufficient), digital offset (typically at -0.900) to optimize the appearance of EB3-dTomato comets and minimize background noise, pinhole at 2 µm, pixel resolution of 512 x 512, and pixel dwell time of 3.15 µs.
7. Acquire a z-stack of the whole embryo with 1 µm section intervals to assess areas of microtubule growth in the whole organism (Figure 1B).
8. Use the 3D z-stack image from step 3.7 to identify regions of interest (ROIs) for EB3-dTomato tracking experiments. Increase the zoom to 3x and draw a rectangular ROI around the specific subcellular area of interest.
9. Acquire a time-lapse movie of a single z-plane using the typical values of imaging parameters: 561 nm laser power at 5%, digital offset of -0.900, pixel resolution of 512 x 512, pinhole at 3 µm, pixel dwell time of 3.15 µs, zoom of 3x, time interval of 500 ms.
   NOTE: 120 time frames will provide a tracking movie of 1 min and should be sufficient for data analysis. The acquisition can continue for longer, provided that bleaching of the fluorophore is minimal (Figure 1C).
10. To activate PST-1P, switch to a 405 nm laser and acquire another time-lapse movie with the 405 nm laser set to 10% power, pixel resolution of 512 x 512, pinhole opened maximally, pixel dwell time of 3.15 µs, zoom of 3x, time interval of 500 ms, and a total of 20 frames (Figure 1D).
11. Switch back to 561 nm laser and repeat acquisition as in step 3.9 to confirm the loss of EB3-dTomato comets after activation of PST-1P (Figure 1E). Ensure that this acquisition takes place as soon as possible following activation.
    NOTE: Step 3.10 can be performed repetitively for longer inhibition of EB3-dTomato comets. However, embryos must be monitored carefully to avoid any harm caused by UV light.
12. To reverse PST-1P back to its inactive trans-state, engage a 514 nm laser at 10% power. Acquire a time-lapse movie with the 514 nm laser set to 10% power, pixel resolution of 512 x 512, pinhole opened maximally, pixel dwell time of 3.15 µs, zoom of 3x, time interval of 500 ms, and a total of 20 time frames (Figure 1F).
13. Repeat step 3.11 to visualize recovery of EB3-dTomato comets (Figure 1G).

4. Image data analysis

1. For analyzing and quantifying the inhibition of microtubule polymerization by PSTs, use software programs available to researchers to suit their specific needs. Those recommended for use will possess a tracking tool, that can manually or automatically track the movement, direction, and speed of EB3-dTomato comets^16,18.

Figure 1: Schematic representation of PST-1P photoactivation and deactivation in the live 3D preimplantation mouse embryo. All experiments are performed in complete darkness (black background) or only by red light illumination. (A) Live preimplantation mouse embryos expressing EB3-dTomato are cultured to 16-cell stage and then transferred to a droplet of KSOM containing 40 µM PST-1P in an imaging chamber slide. (B) A 3D image of the whole embryo allows the assessment of microtubule growth by visualizing the distribution of EB3-dTomato comets. (C) To start the experiment, EB3-dTomato comets are tracked in a subcellular region using time-lapse imaging. (D) Subsequent PST-1P photoactivation in the same subcellular region using a 405 nm laser results in the loss of EB3-dTomato comets (E). Intensified PST-1P activation can be implemented, if necessary, by sequential 405 nm light illumination. (F-G) To reverse PST-1P back to its inactive state and restore EB3-dTomato comets, a 514 nm laser is applied to the same subcellular region. If required, multiple rounds of photoactivation and deactivation can be performed. Please click here to view a larger version of this figure.

Representative Results

In line with the protocol, preimplantation mouse embryos were microinjected with cRNA for EB3, tagged with red fluorescent dTomato (EB3-dTomato). This enables the visualization of growing microtubules as EB3 binds to polymerizing microtubule plus ends²⁴.

The experiments were performed 3 days post-fertilisation (E3) when the mouse embryo is comprised of 16 cells. Any other preimplantation developmental stage can be used, depending on the scientific question to be investigated. To demonstrate minimal photobleaching using the 405 nm laser light activation protocol, a control experiment with an untreated embryo is shown (Figure 2A i). An ROI with striking EB3-dTomato density is chosen based on the acquired 3D image of the embryo. In this example, a single z-plane of an entire cell is used for EB3-dTomato high-resolution imaging. However, different regions of the embryo can be selected as required (for example, cortical region, perinuclear region, a mitotic cell, an inner or outer cell). An image of EB3-dTomato expression before 405 nm laser light illumination shows a highly dense EB3-dTomato signal within the entire cytoplasm of the cell, except the nuclear area (Figure 2A ii). No change in EB3-dTomato density was observed following 405 nm laser light illumination (Figure 2A iii and iv). Thus, the 405 nm laser light per se does not cause any photodamage to the embryo or microtubule dynamics.

Next, 16-cell stage embryos were treated with 40 µM PST-1P 3 h before live imaging and kept in the dark. To successfully perform PST experiments on living 3D organisms, it is critical to find the optimal equilibrium of PST concentration, incubation time, and required laser power to avoid noxious effects¹¹. Under strict dark conditions, strong EB3-dTomato expression can be detected in the 3D image (Figure 2B i), similar to control embryos (Figure 2A i). Thus, PST-1P does not elicit inhibition of microtubule polymerization under dark conditions. Time-lapse imaging of a single z-plane confirmed strong EB3-dTomato expression and microtubule polymerization prior to 405 nm laser light illumination (Figure 2B ii). Within seconds, a reduction of EB3-dTomato signal was observed after 405 nm laser light activation of PST-1P (Figure 2B iii). This loss is reported to last approximately 2 to 20 min before gradually reverting due to thermal relaxation or diffusion into surrounding areas¹¹. Thus, a repetitive 405 nm laser light activation might prolong the duration of EB3-dTomato loss (Figure 1E-D). Importantly, embryos microinjected with EB3-dTomato and incubated with PST-1P show normal embryonic potential, developing to the blastocyst stage (Figure 3, and Supplementary Figure 1), and normal microtubule dynamics, for instance, during mitosis (Supplementary Figure 2). Therefore, inactive PST-1P does not show any toxicity to the embryo.

Controlled PST-1P deactivation can be achieved by 514 nm laser light illumination (Figure 2B iv). Whereas this approach can be useful to determine the origin of microtubule growth after PST-1P-induced growth inhibition, it excludes the use of green fluorescent proteins for live imaging. While it is strongly recommended to work in dark conditions, if PSTs are inadvertently activated, green light illumination enables PST reversion to an inactive trans-state. In this circumstance, an adequate period of recovery for the cells is required as a precautionary measure to ensure cells can return to their pre-activation state. If time permits, it would ideally be a minimum of a few hours.

Figure 2: Live imaging of EB3-dTomato before and after UV light activation and deactivation of PST-1P in the living mouse embryo. (A i) Maximum projection 3D z-stack of an untreated control 16-cell stage mouse embryo labeled for EB3-dTomato. A cytokinetic microtubule bridge is present (denoted by grey arrowhead) alongside multiple interphase bridges (denoted by yellow arrowheads). (A ii) Single z-plane of one cell imaged before 405 nm laser exposure. (A iii-iv) Single z-plane of one cell imaged immediately (A iii) and 4 min (A iv) after 405 nm laser light exposure (indicated by a violet lightning bolt), showing no change in EB3-dTomato signal. (B i) Maximum projection 3D z-stack of a 16-cell stage embryo labeled for EB3-dTomato, which is treated with 40 µM PST-1P and kept in the dark. Interphase bridges are indicated by yellow arrowheads. (B ii) Single z-plane of one cell imaged pre-activation. (B iii) Single z-plane of one cell imaged immediately following 405 nm laser light exposure (indicated by a violet lightning bolt), showing a marked reduction in EB3-dTomato signal. (B iv) Single z-plane of one cell imaged immediately following 514 nm laser light exposure (indicated by a green lightning bolt), showing recovery of EB3-dTomato signal. Nuclei are indicated by the dashed blue line. Scale bars: 15 µm for entire 3D embryos; 5 µm in insets. Please click here to view a larger version of this figure.

Figure 3: Inactive PST-1P culture conditions allow normal embryonic development to blastocyst stage. (A) Single differential interference contrast (DIC) z-plane of an embryo at blastocyst stage (E4.5) following culture in 40 µM PST-1P in the dark. (B) Maximum projection 3D z-stack of the embryo shown in (A), showing EB3-dTomato expression. Scale bars: 10 µm. Please click here to view a larger version of this figure.

In the acquired time-lapse movies, EB3-dTomato labeling appears as "comet-like" structures and enables tracking of growing microtubule filaments (Figure 4). In PST-1P treated embryos kept in dark conditions (Figure 4A), individual EB3-dTomato comets (denoted by white and yellow arrowheads in Figure 4B) can be seen emanating from the interphase bridge, which acts as a non-centrosomal microtubule-organizing center in the early mouse embryo¹⁶ (Figure 4B). A projection of time points (t-stack) depicts the overall number and distance traveled by EB3-dTomato comets (Figure 4B), showing the highest density of EB3-dTomato in the region of the interphase bridge. Following activation of PST-1P with a 405 nm laser light, EB3-dTomato comets can no longer be seen at the base of the interphase bridge and are also absent in the t-stack (Figure 4C).

Figure 4: Tracking of EB3-dTomato comets before and after subcellular UV light activation of PST-1P in the living mouse embryo. (A) Maximum projection 3D z-stack of a 16-cell stage mouse embryo labeled for EB3-dTomato, treated with 40 µM PST-1P and kept in the dark. (B) Four selected single z-planes and corresponding t-stack of an ROI at the interphase bridge (indicated by *) before 405 nm laser illumination, showing EB3-dTomato comets over time. The movement of individual comets is indicated by the yellow and white arrowheads while the color-matched dashed arrowheads show comet positions at previous time points. Times shown are in seconds (s). (C) Four selected single z-planes and corresponding t-stack of the interphase bridge imaged immediately following 405 nm laser light exposure (indicated by a violet lightning bolt) show a loss of EB3-dTomato comets. Scale bars: 10 µm for full 3D embryo; 2 µm for single z-planes; 1.5 µm for t-stack. Please click here to view a larger version of this figure.

Taken together, these results demonstrate the application of PST-1P in the living preimplantation mouse embryo. PST-1P effectively inhibits microtubule polymerization following activation with a 405 nm laser and this inhibition is reversible upon 514 nm laser light illumination on the cellular and subcellular level.

Supplementary Figure 1: Preimplantation embryos cultured in inactive PST-1P demonstrate normal developmental potential to blastocyst stage. (A) Untreated controls and (B) PST-1P treated embryos showing rate of development to blastocyst stage. All embryos were kept in dark conditions throughout development to ensure inactivity of PST-1P. The rate of development to blastocyst stage was comparable between controls (100%, n = 13) and PST-1P treated embryos (91.3%, n = 23). Scale bars are 20 µm. Please click here to download this File.

Supplementary Figure 2: Preimplantation embryos cultured in inactive PST-1P demonstrate normal spindle dynamics during mitosis. (A) Untreated controls and (B) PST-1P treated embryos dividing from 8-cell to 16-cell stage, injected with EB3-dTomato and Histone 2B (H2B)-GFP. All embryos were kept in dark conditions throughout development to ensure inactivity of PST-1P. Alignment of chromosomes can be seen (A i and B i), followed by early anaphase (A ii and B ii), late anaphase (A iii and B iii), and cytokinesis (A iv and B iv) without defects. Scale bars are 10 µm in 3D images and 5 µm in insets. (C) Anaphase lasted a mean of 8 min 53 s in controls and 8 min 40 s in PST-1P treated embryos (p-value = 0.8241) and (D) chromosomes decondensed during telophase a mean of 19 min 38 s in controls and 18 min 51 s following anaphase initiation in PST-1P treated embryos (p-value = 0.2743). For analysis, n = 10 in controls and n = 12 in PST-1P treatment. Furthermore, all cells are divided in one synchronous wave, as expected. Please click here to download this File.

Discussion

The microtubule network is integral to the fundamental inner workings of a cell. Consequently, this presents challenges in manipulating microtubule dynamics in living organisms, as any perturbation to the network tends to have widespread consequences for all aspects of cellular function. The emergence of photoswitchable microtubule-targeting compounds presents a way to precisely manipulate the cytoskeleton at a subcellular level, with superior control for the induction and reversal of microtubule growth inhibition⁹. This protocol demonstrates how microtubule polymerization can be inhibited in the preimplantation mouse embryo by light-activated PST-1P on a subcellular scale, and in a temporally precise manner. Additionally, this inhibition can be reversed to allow investigation of short-term perturbation of the microtubule network.

The protocol presented is optimized for the application of PSTs on live preimplantation mouse embryos and can be used as a foundation for extending their use to various other culture systems. New users of PSTs may encounter some points in their applications requiring troubleshooting, for which several problem-solving steps are presented.

Cytotoxicity of PSTs should be assessed and a working concentration that elicits strong inhibition of microtubule growth with minimal toxicity to cells should be selected. An extensive protocol for optimizing the working concentrations of PSTs and other photoswitches was detailed in a recent pre-print¹¹. In short, new users of PSTs can assess cytotoxicity in dark conditions with serial dilutions of the drug. In this setting, there should be no measurable effect on EB3-dTomato comets and no toxicity to the cells. This provides a starting point for testing a non-cytotoxic concentration under lit conditions. For further experimentation, the lowest concentration which elicits a reduction in EB3-dTomato comets is recommended.

As with any drug treatment, PSTs are capable of being metabolized by the cell; however, from a practical perspective, this is likely to be negligible as demonstrated by incubation of C. elegans embryos in PST-1P for 150 min and subsequent effective isomerisation⁹. For highly metabolically active systems, an increased concentration of PST-1P may be required. PSTs can also diffuse through the culture medium, reducing the load of active PSTs within the ROI. To mitigate this, repeated photoactivation may be necessary, the cytotoxicity of which would need to be assessed on an individual cell-type basis as some systems may be susceptible to phototoxicity. Likewise, activated PSTs may diffuse out of the light-activated ROI and affect surrounding microtubules. However, due to their inherent tendency to relax back to the inactive trans-state, the effect on surrounding areas should be negligible.

Another potential obstacle for the use of PSTs relates to their application in organoids or larger organisms due to the nature of limited light penetrability of these tissues. Larger tissues may not be accessible for the photoactivation of deeper cells. The preimplantation mouse embryo presents a unique challenge in that it is surrounded by an outer coating, the zona pellucida, which can hinder the permeability of drugs added to the media. To mitigate this, embryos are pre-incubated for at least 1 h in PST-1P before imaging. Similar optimization approaches may be applicable to enhance penetration of the drug when using PSTs in whole-organism or organoid systems, such as the addition of acetonitrile to aid cellular uptake of PSTs⁹.

Due to the green light reversibility of PSTs, these compounds are not compatible with imaging of fluorophores requiring excitation below 530 nm (such as GFP). However, users can implement fluorophore tracking with any fluorophore requiring an excitation above 530 nm, for instance, RFP, mCherry, or mPlum. If double-labeling of subcellular structures is required, users can employ red and far-red based fluorophores. If GFP compatibility is essential, a new class of photoswitches, SBTubs, might be of interest²⁵. These compounds are also activated by UV light; however, they are unresponsive to green light, rendering them functionally reversible only by diffusion, but also compatible with all fluorophores with an excitation wavelength of 488 nm or above. Together with PSTs, which offer reversibility but are not compatible with imaging of GFP, these compounds offer the user increased flexibility to take a customized approach and adapt the available photoswitches to their research question.

Conventional small-molecule inhibition, knockout and knockdown approaches can be implemented at whole-cell or whole-organism scales to characterize microtubule action. These approaches provide potent, yet broad and destructive effects on microtubules that are not easily reversible and are often difficult to target in a subcellular or temporally controlled manner. One strategy to circumvent these limitations is through the genetic engineering of a light-inducible optogenetic system. The applications of these systems vary widely and include light-induced heterodimerization, homodimerization, and dissociation of proteins or protein domains²⁶. Recently, an optogenetically controlled end binding protein 1 (EB1) construct was designed, where a light-inducible switch enabled disassociation of the amino- and carboxyl-terminals of the microtubule plus End Binding partner EB1. This rendered the protein functionally inactive with sub-second response times, blocking the growth of microtubules in a human lung carcinoma cell line²⁷. Similarly, an optogenetic switch was used to cross-link microtubule and actin filaments in D. melanogaster to understand the influence of cytoskeletal cross-linking on morphological changes to the cell²⁸. The use of optogenetic tools to manipulate the early embryo is a highly topical area of research and has seen some recent advances¹. An optogenetically modified nuclear export system was developed to induce the mislocalization of nuclear proteins. This could mimic loss-of-function genetic models, with the added flexibility to be conducted at precise developmental time points in D. melanogaster embryos²⁹. Optogenetic approaches are a powerful tool for the spatiotemporal control of microtubule dynamics; however, they can be difficult to engineer, time-consuming, and costly. In contrast, PSTs offer a chemical alternative to complex genetic manipulation that boasts similar response times, specificity, and reversibility with the added advantage of easy use and accessibility.

Another targeted approach to mechanically disrupt microtubule filaments is laser ablation using a two-photon laser. This was applied successfully to ablate the interphase bridge in preimplantation mouse embryos¹⁶. When implemented well, users can achieve a high pay-off. However, executing this technique is challenging and requires a high degree of precision to avoid the destruction or lysis of any surrounding structures or the cells themselves. Furthermore, users may need to repeatedly ablate their samples to achieve the intended effect, as microtubules can regrow quickly³⁰. In contrast, PSTs directly inhibit microtubule polymerization³¹ and avoid undesired effects that may arise with laser ablation, due to their specificity for binding to β-tubulin. Moreover, PSTs can be applied to the whole organism, while laser ablation by nature can only be targeted to a specific region.

Using this protocol, the possibilities for the application of PSTs are wide-ranging and compelling. PSTs can allow the precise, subcellular inhibition of microtubule growth within sub-second response times to study aspects of chromosome segregation during mitotic or meiotic cell division. Trafficking of intracellular cargo could be disrupted by inhibiting the pathways for transport provided by the microtubule network to study endosomal trafficking, movement of mitochondria, or nuclear positioning. PSTs were also used to successfully inhibit microtubule growth in 3D multicellular spheroids grown from human melanoma cell lines^32,33. Coupling the flexibility of PST activation with their superb response times offers researchers a dynamic tool to investigate microtubule dynamics and manipulate cellular function. In addition, the reversibility of PSTs makes their application ideal for the study of both short-term and long-term consequences of microtubule inhibition on a single cell or selected cells within a tissue or organism.

Materials

Name	Company	Catalog Number	Comments
Aspirator tube	Sigma-Aldrich	A5177	For mouth aspiration apparatus
Chamber slides - LabTek	Thermo Fisher Scientific	NUN155411
cRNA encoding for EB3-dTomato	N/A	N/A	Prepared according to manufacturers instructions using mMessage in vitro Transcription kit
Culture dishes - 35mm	Thermo Fisher Scientific	150560
Human chorionic growth hormone	Sigma-Aldrich	C8554
Human Tubal Fluid (HTF) medium	Cosmo-Bio	CSR-R-B071
Imaris Image Analysis Software	Bitplane
Immersion Oil W 2010	Carl Zeiss	444969-0000-000	For use with microscope immersion objective
LED torch - Red light	Celestron	93588
M2 medium	Sigma-Aldrich	M7167
Mice - wild-type FVB/N, males and females	N/A	N/A	Females 8-9 weeks old. Males 2-6 months old.
Microcapillary Pipettes - Kimble	Sigma-Aldrich	Z543306	For mouth aspiration apparatus
Microinjection buffer	N/A	N/A	5 mM Tris, 5 mM NaCl, 0.1 mM EDTA, pH 7.4
Mineral oil	Origio	ART-4008-5P
mMessage In vitro Transcription kit	Thermo Fisher Scientific	AM1340
NanoDrop Spectrophotometer	Thermo Fisher Scientific
Potassium Simplex Optimised Medium (KSOM) medium	Cosmo-Bio	CSR-R-B074
Pregnant mare serum gonadotrophin	Prospec Bio	HOR-272
PST-1P	N/A	N/A	Borowiak, M. et al., Photoswitchable Inhibitors of Microtubule Dynamics Optically Control Mitosis and Cell Death. Cell. 162 (2), 403-411, doi:10.1016/j.cell.2015.06.049, (2015).
RNA purification kit	Sangon	B511361-0100
Ultrapure water	Sigma-Aldrich	W1503
ZEN Black Software	Carl Zeiss