A single-component optogenetic toolkit for reversible visualization and programmable control of microtubule dynamics

Summary

Microtubules form dynamic cytoskeletal scaffolds essential for intracellular transport, organelle positioning, and spatial organization of signaling. Their architecture and function are continuously remodeled through the concerted actions of microtubule-associated proteins (MAPs), post-translational modifications (PTMs), and molecular motors. To precisely interrogate these processes in living systems, we developed a genetically encoded, single-component optogenetic platform for spatiotemporal control of microtubule organization and dynamics. By harnessing light-induced oligomerization to regulate microtubule association, this system supports reversible microtubule labeling and plus-end tracking, localized control of tubulin PTMs, optically regulated kinesin-driven cargo transport, and inducible microtubule severing within a unified design strategy. Using these tools, we reveal how local microtubule integrity governs lysosomal trafficking and endoplasmic reticulum (ER)-associated signaling dynamics. Collectively, this modular optogenetic toolkit bridges molecular design with cytoskeletal function, offering a versatile platform to dissect how dynamic cytoskeletal architectures coordinate intracellular organization, transport, and signaling.

Graphical abstract

Highlights

• •CRY2 oligomerization supports single-component design for optogenetic MT control
• •OptoMT and OptoTIP provide reversible microtubule labeling and plus-end tracking
• •Modular design enables light-controlled tubulin PTMs and OptoMotor cargo transport
• •OptoSAW locally disrupts MTs to probe signaling and organelle trafficking

Motivation

Microtubules (MTs) are dynamic, polarized cytoskeletal networks whose functions are shaped by MT-associated proteins (MAPs) and regulators of tubulin post-translational modifications (PTMs), which together confer spatial and temporal plasticity. Defining how local MT remodeling drives cellular function remains challenging because conventional genetic, pharmacological, and labeling approaches can perturb MT dynamics, lack reversibility or subcellular precision, and are often incompatible with long-term live-cell imaging. Existing optogenetic tools enable control of selected MT-associated processes but typically rely on multi-component designs or distinct engineering strategies. Here, we develop a single-component, modular optogenetic platform that uses light-induced clustering to reversibly visualize, modify, transport along, and remodel MTs with spatiotemporal control.

Lan et al. develop a single-component optogenetic toolkit that uses light to visualize, modify, transport along, and remodel microtubules in living cells. This modular platform links local cytoskeletal changes to organelle positioning and signaling, providing a flexible way to study how cell architecture controls function.

Introduction

Microtubules (MTs) form a dynamic and polarized cytoskeletal network that supports a broad spectrum of cellular processes, from directed intracellular transport and cell division to signal transduction and organelle positioning.^1,2 These diverse functions depend on the coordinated action of MT-associated proteins (MAPs) that encompass molecular motors, adapter proteins, severing enzymes, and regulators of post-translational modifications (PTMs), which together endow the MT cytoskeleton with remarkable spatial and temporal plasticity.^3,4 While decades of biochemical and structural studies have delineated many of the core mechanisms underlying MT assembly and turnover,^5,6 they often fall short of capturing the highly localized and transient regulation that occurs in living cells, where MAPs tend to operate in concert and context dependently.

Conventional methods for visualizing and manipulating MTs, such as tubulin overexpression, immunostaining, or chemical labeling, often perturb polymerization dynamics or are incompatible with live-cell imaging and lack reversibility.^7,8,9 Over the past decade, optogenetic approaches have emerged as powerful strategies for controlling cytoskeletal dynamics with spatiotemporal precision.¹⁰ Multiple tools have been developed to regulate MT behavior, including optogenetic end-binding protein 1 (EB1) dissociation,^11,12 opto-katanin-mediated MT severing,¹³ light-controlled motor recruitment,^14,15,16,17 and photoswitchable MT-targeting drugs.⁹ Together, these tools demonstrate that light-dependent and reversible interactions with MT lattices or plus-ends can be harnessed to interrogate MT behavior and MAP function with subcellular resolution and, in principle, can be extended to support localized recruitment of regulatory or enzymatic activities.

Building on these exciting advances, we sought to explore a complementary single-component optogenetic approach that departs from existing multi-component designs, diversifies optogenetic engineering strategies, and expands the kinetic and functional space available for MT interrogation. Here, we present a design strategy that exploits light-induced oligomerization to convert weak MT-binding motifs or MAP fragments into reversible MT engagement or regulatory modules. Rather than designing distinct optogenetic systems for each facet of MT regulation, we leverage light-controlled avidity as a shared design logic to visualize and manipulate multiple layers of MT biology within a unified engineering framework. Collectively, this work illustrates a modular, single-component optogenetic strategy that complements existing multi-component systems and provides a unified design approach for dissecting the structural, transport, and remodeling mechanisms that underpin MT organization, thereby bridging molecular interactions to cellular-scale dynamics and providing additional avenues for exploring cytoskeletal function in physiology and disease.

Results

Design of optogenetic probes for reversible labeling of the MT cytoskeleton

To achieve reversible MT labeling with minimized perturbation to its dynamics, we selected the N-terminal photolyase homology region of Arabidopsis thaliana cryptochrome 2 (CRY2-PHR; residues 1–498)^{18,19,20,21,22,23,24,25} as the core photosensory module. CRY2-PHR undergoes robust blue-light-induced oligomerization in living cells without requiring an exogenous chromophore. We reasoned that light-induced CRY2 clustering could increase the avidity of weak tubulin-binding motifs, thereby shifting their binding equilibrium toward stronger MT association in a light-dependent manner (Figure 1A). We generated a panel of chimeric constructs by fusing CRY2 to tubulin-binding domains derived from four well-characterized MT binders, including kinesin (KIF5A), cytoplasmic linker region of 170 kDa (CLIP170), EB1, and calmodulin-regulated spectrin-associated protein 1 (CAMSAP1) (Figure 1B).^{1,26,27,28,29,30,31,32} In the dark, the designed hybrid protein is anticipated to be evenly distributed in the cytosol, whereas blue light stimulation at 470 nm could trigger rapid and reversible tracking of the MT network (Figure 1A). Among all tested constructs, CRY2-CLIP170_129–350 (designated “OptoMT”) yielded the least basal activity while exhibiting robust MT labeling, characterized by a high signal-to-noise ratio and excellent reversibility (Figure 1C and Video S1). Localized blue light stimulation of a defined region of interest (ROI) induced confined MT labeling within and immediately adjacent to the illuminated area, demonstrating subcellular spatial precision (Figure 1D). OptoMT exhibited rapid kinetics, with an activation half-life (t_1/2,_ON) of 10 s and a decay half-life (t_1/2,_OFF) of 210 s following cessation of illumination, making it suitable for dynamic and reversible live-cell imaging (Figure 1E; Table 1). This reversibility is biologically important because it permits repeated measurements in the same living cell, enables direct comparison of MT architecture before, during, and after a defined optical stimulus, and reduces cumulative perturbation from persistent MT binding, thereby supporting long-term live-cell imaging without reliance on fixation or chemical labeling. Labeling specificity was further validated by immunostaining in HeLa cells expressing mCherry (mCh)-OptoMT with an anti-α-tubulin antibody. Fixed-cell analysis revealed tight colocalization of OptoMT with the endogenous MT network (Figure 1F).

Figure 1.

OptoMT for photo-inducible visualization of microtubule and mitosis in living cells

(A) Schematic of OptoMT design and its light-induced association with MT.

(B) Domain architecture of OptoMT variants. The photosensitive module CRY2 (aa 1–498) was fused to tubulin-binding domains derived from Kinesin, EB1, CLIP170, or CAMSAP1. The optimal construct (variant 3, highlighted in red) exhibited minimal basal activity in the dark but strong MT binding upon blue light stimulation.

(C) Confocal images of HeLa cells expressing the indicated mCh-OptoMT variants with or without blue light exposure (indicated by blue bars). Right, quantification of normalized MT-to-cytosol fluorescence intensity ratio before and after 1 s external blue light stimulation (470 nm, 40 μW/mm²). n = 24 cells from three independent biological replicates. Also see Video S1.

(D) Confocal images showing precise spatiotemporal control of OptoMT labeling of MTs within the regions indicated upon pulsed blue light exposure (built-in 488-nm excitation channel, 1% laser output).

(E) Confocal images showing robust and reversible OptoMT labeling of MTs across two successive dark-light cycles (external, 470 nm, 30 s, 40 μW/mm²).

(F) Confocal images of HeLa cells expressing mCh-OptoMT (red) co-stained with anti-α-tubulin (green) and DAPI (blue). Cells were either kept in the dark (top) or illuminated with external blue light (470 nm, 40 μW/mm², 30 s) before fixation and immunostaining.

(G) Live-cell imaging of HeLa cells co-expressing GFP-OptoMT (green in the merged image) and H2B-mCh (red), showing MT cytoskeleton and mitotic progression at different cell cycle stages before and after external blue light exposure (470 nm, 40 μW/mm², 5 s). Also see Video S2.

(H) Confocal images of L3-stage C. elegans expressing GFP::α-tubulin and mCh::OptoMT in epithelia. Blue light (external, 470 nm, 30 s, 40 μW/mm²) induces MT binding by mCh::OptoMT.

(I and J) Confocal images (I) and quantitative analysis (J) of OptoMT-mediated reversible MT labeling in C. elegans. The acquired data points were fitted by a single exponential decay function (t_1/2,_on = 12.4 ± 3.2 s; t_1/2,_off = 252 ± 31 s).

Scale bars, 5 μm. Error bars denote SEM.

Table 1.

Summary of activation and deactivation half-lives of tools in the study

Tool name	Version	Components	FMT/Fcytosol or Fcomet/Fcytosol		ON phase (t1/2,, s)	OFF phase (t1/2, s)
			Dark	Light
OptoMT	v.1	CRY21–498-KIF5A1–330	1.1	3.5	–	–
	v.2	CRY21–498-EB11–191	1.1	3.5	–	–
	v.3	CRY21–498-Clip170129–350	1.0	4.7	10.1 ± 4.2	210.0 ± 28.2
	v.4	CRY21–498-CAMSAP11270–1473	1.7	4.9	–	–
OptoTIP variants	v.1	CRY21–498-APC2786–2824	1.0	3.0	–	–
	v.2	CRY21–498-DST5469–5485	1.0	4.4	11.1 ± 4.7	204.0 ± 43.8
	–	CRY21–498, L348F-DST5469–5485	1.0	3.2	–	∼954
	–	CRY21–498, W349H-DST5469–5485	1.0	4.2	–	∼108
	–	CRY21–498, W349L-DST5469–5485	1.0	4.5	–	∼114
	–	CRY21–498, W349E-DST5469–5485	1.0	3.5	–	∼162
	–	CRY21–498, W349A-DST5469–5485	1.0	1.1	–	–
	–	CRY21–498, W349G-DST5469–5485	1.0	1.2	–	–
	–	CRY21–498, W349S-DST5469–5485	1.0	1.1	–	–
	–	CRY21–498, W349M-DST5469–5485	1.0	1.5	–	–
	–	CRY21–498, W349R-DST5469–5485	1.0	1.2	–	–
	–	CRY21–498, W349D-DST5469–5485	1.0	1.1	–	–
	–	CRY21–498, W349Q-DST5469–5485	1.0	1.4	–	–
	–	CRY21–498, W349T-DST5469–5485	1.0	1.1	–	–
	–	CRY21–498, W349I-DST5469–5485	1.0	1.3	–	–
	v.3	CRY21–498-DST5474–5485	1.0	4.3	–	–
	v.4	CRY21–498-STIM1233–685	1.0	2.6	∼45	∼320
	v.5	CRY21–498-STIM1238–685	1.0	2.9	–	–
	v.6	CRY21–498-STIM1240–685	1.0	2.9	–	–
	v.7	CRY21–498-STIM1245–685	1.0	2.2	–	–
	v.8	CRY21–498-STIM1250–685	1.0	2.3	–	–
	v.9	CRY21–498-STIM1252–685	1.0	2.5	–	–
	v.10	CRY21–498-STIM1258–685	1.0	2.4	–	–
	v.11	CRY21–498-STIM1265–685	1.0	2.3	–	–
	v.12	CRY21–498-STIM1343–685	1.0	3.1	–	–
	v.13	CRY21–498-STIM1443–685	1.0	2.7	–	–
	v.14	CRY21–498-STIM1490–685	1.0	2.2	–	–
	v.15	CRY21–498-STIM1590–685	1.0	2.2	–	–
	v.16	CRY21–498-STIM1620–685	1.0	2.6	–	–
	v.17	CRY21–498-STIM1630–685	1.0	3.0	∼15	∼260
	v.18	CRY21–498-STIM1630–670	1.0	3.4	–	–
	v.19	CRY21–498-STIM1630–660	1.0	3.8	∼12	∼220
	v.20	CRY21–498-LOV2-DST5470–5485	1.0	2.1	9.1 ± 4.1	40.2 ± 6.6
Opto-αTAT1	–	αTAT1-CIBN-NLS-LEXY	–	–	114 ± 15	103 ± 8

The best-performing construct in each tool category is shown in bold.

Video S1. Reversible, light-induced labeling of the MT cytoskeleton by OptoMT, related to Figure 1

Time-lapse confocal imaging of HeLa cells expressing mCh-OptoMT during two repeated dark-light cycles (external, 470 nm, 30 s, 4 mW/cm²). OptoMT rapidly associated with MTs upon illumination and subsequently redistributed uniformly throughout the cytoplasm after light withdrawal, demonstrating reversible binding to the MT network.

We next applied OptoMT to examine MT organization during mitosis. Using H2B as a chromatin marker, light-induced labeling enabled clear visualization of MT dynamics across distinct mitotic stages in live HeLa cells (Figure 1G), without detectable perturbation to cell cycle progression, mitotic division, or viability (Figures S1A–S1C and Video S2). To extend its application in vivo, we co-expressed GFP::tubulin (eGFP::TBB-2 β-tubulin) and mCherry::OptoMT under an epithelial-specific promoter in L3-stage C. elegans. GFP::tubulin revealed circumferential MT bundles in the syncytial epithelium, oriented perpendicular to the body axis (Figure 1H). In the dark, mCherry::OptoMT was diffusely distributed throughout the epidermis, whereas blue light illumination triggered its rapid redistribution into patterned structures that closely colocalized with GFP::tubulin, confirming light-dependent MT labeling in vivo. OptoMT also exhibited robust and reversible MT labeling in C. elegans, with an activation half-life (t_1/2, _ON) of 12 s and a decay half-life (t_1/2, _OFF) of 252 s following light withdrawal (Figures 1I and 1J). Thus, OptoMT recognizes C. elegans as well as human MTs.

Video S2. Monitoring mitosis in HeLa cells, related to Figure 1

Monitoring mitosis in HeLa cells transfected with H2B-GFP alone (as a chromatin marker, bottom panel) or co-transfected with mCh-OptoMT (top panel). No significant perturbation of host cell division was observed following co-expression and light illumination. External blue-light pulses (470 nm, 40 μW/mm²) were applied for 5 s every 2 min.

Together, these results establish OptoMT as a single-component optogenetic probe that enables reversible and light-controlled MT labeling with minimal cellular perturbation, applicable both in vitro and in vivo.

Design of photoswitchable MT plus-end trackers

To achieve reversible tracking of growing MT plus-ends, we adapted the optogenetic clustering strategy to MT tip-localization domains containing the S/T-x-I-P (SxIP) consensus motif.^1,33,34 CRY2 was fused to SxIP-containing fragments of varying lengths derived from three representative MT plus-end tracking proteins (+TIPs), adenomatous polyposis coli (APC), dystonin (DST), and the stromal interaction molecule 1 (STIM1), which engage the EB1 at growing MT ends (Figures 2A and 2B; Figure S2A).²⁹ Among the tested variants, CRY2-DST_5469–5485 (V2, hereafter designated “OptoTIP”) showed the most robust comet-like localization upon blue light illumination (Figure 2C; Figures S2B and S2C). OptoTIP displayed fast activation (t_1/2,_ON = 11 s) and moderate deactivation (t_1/2, _OFF = 204 s) kinetics, which were fully reversible across repeated dark-light cycles (Figure 2C; Table 1; Video S3). When co-expressed with GFP-EB1, mCherry-OptoTIP precisely colocalized with EB1 comets in a blue light-dependent manner, showing nearly identical fluorescence intensity profiles along MT tips (Figure 2D; Figure S2D). Nocodazole treatment, which depolymerizes MTs,^35,36 rapidly abolished the comet-like pattern formation, even under continuous illumination, indicating that OptoTIP labeling is strictly dependent on intact MT polymer assembly (Figure 2C). No significant differences were observed in EB1 comet velocities between mCherry- (control) and mCherry-OptoTIP-expressing cells before and after blue light exposure, demonstrating that OptoTIP does not perturb endogenous MT growth (Figure S2E). Localized blue light stimulation within defined subcellular regions further induced spatially confined comet-like pattern formation, demonstrating the high spatiotemporal precision of this system (Figure 2E).

Figure 2.

OptoTIP as a single-component optogenetic actuator for reversible tracking of microtubule plus-ends (+TIPs)

(A) Schematic illustrating the design and light-induced association of OptoTIP with EB1 to track MT plus-ends.

(B) OptoTIP variants were constructed by fusing CRY2 with EB1-binding motifs derived from APC, DST, or STIM1.

(C) Confocal images and quantification of OptoTIP-mediated MT plus-end tracking during repeated dark-light cycles, followed by nocodazole treatment (20 μM). Blue bars indicate periods of illumination (external, continuous, 470 nm, 40 μW/mm²). Data were fitted with a single exponential decay function (t_1/2, _on = 11.1 ± 4.7 s; t_1/2, _off = 204 ± 44 s). n = 10 cells from three independent biological replicates. Also see Video S3.

(D) Confocal images showing photo-activatable colocalization of mCh-OptoTIP (red) with GFP-EB1 (green) at growing MT plus-ends following pulsed blue light exposure (built-in 488-nm excitation channel, 1% output), demonstrating faithful tracking of EB1-marked comets.

(E) Confocal images showing precise spatiotemporal control of OptoTIP tracking by visualization of successively selected photostimulation regions (built-in 488-nm excitation channel, 1% output).

(F) Confocal images of L3-stage C. elegans expressing GFP::α-tubulin and mCh::OptoTIP, showing blue-light-inducible (external, 470 nm, continuous, 40 μW/mm²) OptoTIP tracking of growing MT plus-ends.

(G) Enlarged region (white boxed inset) from (F) showing reversible OptoTIP relocalization upon alternating dark-light cycles.

(H) Schematic depicting the design of LOV2-based OptoTIP variants. To enable tracking of MT plus-ends with faster deactivation kinetics, the SxIP motif derived from dystonin (DST_5469–5485) was positioned downstream of the C-terminal Jα helix of LOV2, which was then fused either to a tetrameric dsRed (D series; D1–D6) or the light-inducible oligomerization module CRY2 (V20).

(I) Quantification of comet-to-cytosol fluorescence intensity ratios for LOV2-based OptoTIP constructs. mCherry fluorescence at comets and adjacent cytosolic regions was measured before and after 30 s external blue light exposure (470 nm, 40 μW/mm²). n = 17–25 cells from three independent biological replicates.

(J) Confocal images of HeLa cells expressing mCherry-CRY2-LOV2-SxIP (V20, DST_5470–5485) showing light-dependent plus-end tracking with rapid deactivation kinetics during sequential dark-light-dark cycles (external, 470 nm, continuous, 40 μW/mm²).

(K) Quantitative analysis of comet-to-cytosol fluorescence intensity ratios for the mCherry-CRY2-LOV2-SxIP (V20) construct. The measured kinetics revealed efficient MT plus-end tracking with an activation half-life of 9.1 ± 4.1 s and deactivation half-life of 40.2 ± 6.6 s. Data were fitted with a single exponential decay function. n = 16 cells from three independent biological replicates.

Scale bars, 5 μm. Error bars denote SEM.

Video S3. Reversible light-controlled tracking of MT plus-ends by OptoTIP, related to Figure 2

In HeLa cells, mCh-OptoTIP exhibited reversible comet-like accumulation at MT plus-ends under repeated dark-light cycles. Kinetic analysis revealed an activation half-life (t_1/2, ON) of around 11 s and a decay half-life (t_1/2, OFF) of about 204 s following light withdrawal.

OptoTIP was robustly expressed across diverse cell types, including a dozen cell lines derived from both excitable and non-excitable tissues, enabling light-inducible tracking of MT plus-ends in multiple cellular contexts (Figure S2F). To evaluate its performance in vivo, we co-expressed GFP::tubulin and mCherry::OptoTIP in the epidermis of C. elegans larvae, as described earlier. Blue light illumination triggered robust and reversible tracking along GFP-labeled MT bundles on both dorsal and ventral epidermal surfaces (Figures 2F and 2G). Together, these results establish OptoTIP as a single-component optogenetic probe that enables light-inducible tracking of growing MT plus-ends in real time in living cells and animals without appreciable perturbation of endogenous cytoskeletal dynamics.

To expand the kinetic range of OptoTIP, we introduced photocycle mutations in CRY2 at residues L348 and W349,³⁷ located near the flavin adenine dinucleotide (FAD)-binding site (Figure S3A). The L348F substitution produced a long-lived variant (t_1/2, _OFF = 15.9 min), whereas W349 substitutions (e.g., W349H, W349L, and W349E) either accelerated decay or abolished tip-tracking activity (Figure S3B). To achieve faster deactivation, we adopted LOV2 from oat phototropin, a blue light photosensory domain with a shorter photocycle.^{38,39,40,41,42,43} We placed the SxIP motif immediately downstream of the C-terminal Jα helix of LOV2, reasoning that EB1 binding would be sterically blocked in the dark and unmasked upon blue-light-induced unfolding of the Jα helix (Figure 2H; Figure S3C). We further fused LOV2-SxIP constructs to either dsRed (a constitutive tetramer) or CRY2 (a light-induced oligomer) to enhance avidity. After optimizing linker regions and the SxIP motif length (Figure 2I; Figure S3D), we identified a LOV2-based OptoTIP variant (V20, Figure S3D) that exhibited rapid and reversible cytosol-to-comet transitions under repeated dark-light cycles (Figures 2J and 2K). This construct substantially accelerated deactivation kinetics, reducing the decay half-lives from minutes to approximately 40 s (Figure 2K; Table 1). Together, these efforts yielded a toolkit of CRY2- and LOV2-based OptoTIP variants that enable reversible and light-tunable tracking of growing MT plus-ends, with kinetic profiles ranging from seconds to minutes.

Optogenetic manipulation of tubulin PTMs

We next explored whether OptoTIP or OptoMT could serve as modular platforms for light-inducible PTMs of tubulin. As a proof of concept, we focused on α-tubulin acetyltransferase (αTAT1), which catalyzes the acetylation of α-tubulin at the position K40 to enhance MT resistance to mechanical and chemical stress.⁴⁴ To render αTAT1 light responsive, we incorporated the LOV2-based light-inducible nuclear export system (LEXY)⁴⁵ and two nuclear localization signals (NLSs), yielding Opto-αTAT1. In the dark, the NLS sequences confined the enzyme within the nucleus, whereas blue light triggered LEXY-mediated export to the cytoplasm (Figure 3A; Figure S4A). For MT-specific recruitment, the N-terminal domain of CIB1 (CIBN, residues 1–81), which heterodimerizes with CRY2 under photostimulation,^37,46,47 was appended to Opto-αTAT1 (Figure S4A). Upon blue light illumination, a portion of nuclear Opto-αTAT1 translocated to the cytosol (t_1/2, _ON = 1.9 min; t_1/2, _OFF = 1.7 min) and was subsequently recruited to OptoTIP- or OptoMT-decorated MT filaments through CRY2-CIBN heterodimerization (Figure 3B; Figures S4B and S4C). For these long-term PTM induction experiments, we used pulsed rather than continuous blue light illumination to minimize total light exposure, photobleaching, and potential phototoxicity. Given the relatively slow dark-state recovery of Opto-αTAT1, OptoTIP, and OptoMT, intermittent illumination was sufficient to maintain light-induced cytosolic translocation and MT recruitment while reducing light-associated perturbations during prolonged stimulation. This relocalization induced a robust increase in tubulin acetylation, as evidenced by enhanced acetyl-α-tubulin immunostaining (Figures 3C and 3D) and the progressive appearance of acetylated bands in western blots after illumination (Figure 3E). In contrast, control cells lacking Opto-αTAT1 displayed only basal acetylation levels (Figure 3D).

Figure 3.

OptoMT or OptoTIP enables light-inducible post-translational modifications of tubulin

(A) A simplified schematic depicting the design of photosensitive acetyltransferase αTAT1, Opto-αTAT1. Upon blue light stimulation, Opto-αTAT1 undergoes nucleus-to-cytosol translocation and is further recruited toward MT via light-dependent interaction with OptoTIP or OptoMT to efficiently catalyze the acetylation of α-tubulin. See Figure S4 for details of the photosensitive modules.

(B) Light-inducible nuclear export of mCh-Opto-αTAT1 with subsequent recruitment to MT in HeLa cells co-expressing GFP-OptoTIP. Upper: confocal images of the same cell before and after exposure to pulsed external blue light at 470 nm (40 μW/mm², 3 s on and 30 s off; blue bar). Lower: normalized cytosolic mCherry signals of Opto-αTAT1 during two repeated light-dark cycles. n = 24 cells from three independent experiments.

(C) Confocal images of HeLa cells co-expressing Opto-αTAT1 and OptoTIP after varying durations of pulsed external blue light exposure (470 nm, 40 μW mm⁻², 3 s on and 30 s off), followed by fixation and immunostaining for acetylated α-tubulin.

(D) Quantification of α-tubulin acetylation levels from images in (C). Data represent 102 cells per time point from three independent biological replicates.

(E) Immunoblot analysis of light-triggered α-tubulin acetylation (Ac-α-tubulin) in HeLa cells expressing Opto-αTAT1. Right, densitometric quantification showing the increased Ac-α-tubulin/GAPDH ratio upon blue light illumination. n = 3 independent biological replicates.

(F) Time-lapse confocal imaging of MT cytoskeleton (probed by GFP-OligoMT) in HeLa cells co-expressing OptoMT and Opto-αTAT1 in the presence of 2 μM nocodazole. Top (gray bar; dark): cells were cultured in the dark. For acquiring GFP signals, cells were very briefly subjected to photostimulation for 1–2 s without eliciting nuclear export of Opto-αTAT1. Bottom (blue bar; light): cells were exposed to external blue light (470 nm, 40 μW/mm²) before imaging with a pulse of 3 s on and 30 s off.

(G) Quantification of the MT-over-cytosol fluorescence intensity ratio over 120 min of nocodazole treatment. n = 28 (dark) and 30 (light) cells from three independent biological replicates.

(H) The bar graph and scatter dots showed the averaged values and distribution of MT-to-cytosol intensity ratios of GFP in cells cultured in the dark or under blue light illumination before and after nocodazole treatment. n = 28 (dark) and 30 (light) cells from three independent biological replicates.

Each symbol represents the average of 5–6 measurements from one cell. ∗∗∗p < 0.001 (two-tailed Student’s t test).

(I–K) Light-dependent tubulin detyrosination enabled by Opto-VASH.

(I) Confocal images of HeLa cells co-expressing Opto-VASH and OptoTIP after varying durations of external blue light exposure (470 nm, pulses of 3 s on and 30 s off, 40 μW/mm²), followed by immunostaining for detyrosinated tubulin (DeY-tubulin).

(J) Quantification of DeY-tubulin levels from images shown in (I). Data represent 96 cells per time point from three independent biological replicates.

(K) Immunoblot analysis of DeY-tubulin in HeLa cells co-transfected with Opto-VASH and OptoTIP with or without blue light stimulation (external, 470 nm, pulses of 3 s on and 30 s off, 40 μW/mm²).

Right, densitometric quantification of DeY-tubulin intensity. n = 3 independent biological replicates.

Scale bars, 5 μm. Error bars denote SEM.

To determine whether light-induced acetylation enhances MT stability, HeLa cells co-expressing GFP-OptoMT and Opto-αTAT1 were treated with nocodazole under illuminated or dark conditions. In the absence of light, nocodazole gradually disassembled the MT network within 90 min (Figures 3F–3H). Photoactivation of Opto-αTAT1, however, markedly increased MT resistance to depolymerization, largely preserving cytoskeletal integrity over the same period (Figures 3F–3H). These results demonstrate that Opto-αTAT1 enables light-dependent α-tubulin acetylation and strengthens the MT network against chemical stress.

Building on this approach, we next engineered a light-inducible detyrosination system, termed Opto-VASH, by fusing vasohibin 1 (VASH1), a tubulin carboxypeptidase that removes the C-terminal tyrosine from α-tubulin,⁴⁸ to the same photosensitive module. When coupled with OptoMT, Opto-VASH progressively increased detyrosinated tubulin levels under pulsed blue light stimulation (3 s ON, 30 s OFF), as confirmed by both immunostaining (Figures 3I and 3J) and western blotting (Figure 3K).

Together, these findings establish OptoTIP and OptoMT as modular scaffolds for temporal control of tubulin PTMs, enabling light-driven modulation of acetylation and detyrosination within the MT network of live cells.

Optogenetic reconstruction of intracellular transport

Kinesin typically consists of an N-terminal motor domain, a central coiled-coil stalk that mediates dimerization and processive movement, and a C-terminal tail domain responsible for cargo binding and specificity^49,50 (Figure 4A; Figure S5A). To render kinesin activity light responsive, we replaced the native coiled-coil stalk and C-terminal tail with CRY2 (Figure 4B), reasoning that blue-light-induced CRY2 oligomerization could reconstitute motor dimerization and activation (Figure 4C).

Figure 4.

OptoMotor designed to photo-manipulate intracellular transport

(A) Domain organization of mouse KIF5A (mKIF5A), showing the N-terminal motor head, central coiled-coil stalk, and C-terminal cargo-binding tail.

(B) Schematic of mKIF5A truncation variants fused to CRY2 for optogenetic activation. Variant V4 (residues 1–379) was identified as the optimal construct and designated OptoMotor.

(C) Design principle of OptoMotor. Blue light illumination (external, 470 nm, continuous, 40 μW/mm²) induces CRY2 oligomerization to drive reassembly of the truncated motor complex and restoring plus-end-directed motility along under MTs.

(D) Confocal images of selected CRY2-KIF5A truncation variants expressed in HeLa cells under dark and illuminated conditions (external, 470 nm, 40 μW/mm²). Variant V1 exhibited constitutive peripheral accumulation, V4 showed robust light-induced redistribution to the cell periphery without appreciable basal activity, and V6 displayed pronounced MT labeling. See Figure S5 for the complete set of variants.

(E) Quantification of periphery-to-cytosol fluorescence ratios across truncation variants. Each symbol represents the mean of 5–6 cells from a single imaging field. A total of 40–50 cells were analyzed per construct across three independent biological replicates.

(F) Time-lapse images showing reversible distribution of OptoMotor during repeated dark-light cycles (external, 470 nm, 40 μW/mm²). Also see Video S4.

(G) Kinetic trace of peripheral intensity changes showing OptoMotor activation (t_1/2, _ON = 0.9 min) and deactivation (t_1/2, _OFF = 3.7 min).

(H) Localized blue light illumination (built-in 488-nm excitation channel, 1% output) within defined regions of interest (ROIs R1, R2) elicited spatially confined redistribution of OptoMotor in single cells.

(I) Quantification of relative fluorescence intensity changes within ROIs confirming high spatial precision of light-triggered activation.

(J) Schematic of the chimeric OptoMotor-Tail construct, generated by fusing the cargo-binding C-terminal tail of mKIF5A (residues 907–1027) to OptoMotor. Upon blue light activation, OptoMotor-Tail drives peripheral transport of lysosomes to enhance mTORC signaling.

(K) Confocal images of HeLa cells co-expressing LAMP1-GFP and mCh-OptoMotor-Tail showing light-induced redistribution of lysosomes toward the cell periphery (built-in 488-nm excitation channel, 1% output).

(L) Immunoblot showing phosphorylation of S6K at T389 (P-S6K) as a readout of mTORC1 activity. Blue light illumination (external, 470 nm, pulses of 3 s on and 30 s off, 40 μW/mm²) enhanced P-S6K levels in cells expressing OptoMotor-Tail, consistent with lysosomal repositioning-mediated activation of mTORC1 signaling. Starved cells before and after nutrient recovery were used as negative and positive controls for reporting mTORC1 activity.

Scale bars, 5 μm. Error bars denote SEM.

Using KIF5A, a kinesin-1 heavy chain with well-characterized biophysical properties,^51,52 as the engineering scaffold, we systematically truncated its coiled-coil region to generate seven CRY2-fused variants with progressively shortened neck linkers (Figure 4B). This series enabled the identification of configurations that preserved processive stepping while supporting blue-light-dependent activation (Figures 4B and 4C). When expressed in HeLa cells, the variants displayed distinct localization patterns before and after light illumination. Constructs retaining longer coiled-coil regions (beyond residue 397; V1–V3) accumulated at the cell periphery even in the dark, suggesting that excessive coiled-coil inclusion promotes spontaneous plus-end targeting (Figures 4D and 4E; Figure S5B). The shorter truncations behaved differently, with KIF5A_1–327-CRY2 (V6) remaining cytoplasmic in the dark yet binding strongly to MTs upon activation (Figure 4D), functioning as a light-inducible MT-binding probe. By contrast, KIF5A_1–359-CRY2 (V5) remained unresponsive to light, while KIF5A(1–317)-CRY2 (V7), which harbors an incomplete motor domain, formed punctate aggregates, likely arising from intrinsic CRY2 clustering (Figure 4E; Figure S5B). Among these, KIF5A_1–379-CRY2 (V4, termed OptoMotor) showed optimal performance, remaining diffusely cytoplasmic in the dark and rapidly redistributing to the cell periphery under blue light (Figures 4D and 4E; Video S4). Quantitative analysis revealed rapid and reversible light-inducible activation, with an activation half-life (t_1/2,_ON) of 0.9 min and a deactivation half-life (t_1/2,_OFF) of 3.7 min (Figures 4F and 4G; Video S4). Localized blue light stimulation further induced region-specific redistribution of OptoMotor, confirming its high spatiotemporal precision and controllability (Figures 5H and 5I). Together, these results establish OptoMotor as a single-component light-inducible hybrid motor capable of reconstituting processive plus-end-directed transport.

Figure 5.

OptoSAW enables precise control of microtubule severing

(A) Schematic illustration of the single-component OptoSAW constructs generated by directly fusing the minimal MT-severing domain of spastin (residues 228–616 carrying the K3/Q3 mutation, ³⁰⁹KKK³¹¹ → QQQ) to either OptoMT or OptoTIP. See Figure S6 for spastin engineering and characterization details.

(B) Time-lapse confocal images monitoring light-inducible MT severing in single HeLa cells expressing mCh-OptoMT-SAW (red) and GFP-OligoMT (green), the latter serving as an MT marker (built-in 488-nm excitation channel, 1% output).

(C) Fluorescence intensity profiles of GFP (green) and mCherry (red) signals measured along the dashed line shown in (B), demonstrating the progressive decline of filament continuity during blue light illumination.

(D) Quantification of filament-associated GFP intensity from images shown in (B). n = 32 cells from three independent biological replicates.

(E) Time-lapse confocal images of HeLa cells co-expressing EB1-GFP (green) and mCherry-OptoTIP-SAW (red), showing light-induced disruption of +TIP comet dynamics (built-in 488-nm excitation channel, 1% output). See Video S5.

(F) Confocal images demonstrating spatially restricted MT severing. Patterned blue light illumination selectively induced MT disassembly within targeted regions (built-in 488-nm excitation channel, 1% output).

(G) Time-lapse imaging of HeLa cells co-expressing mCherry-OligoTIP (+TIP marker) and OptoTIP-SAW following blue light stimulation (built-in 488-nm excitation channel, 10 μW/mm²), revealing progressive disassembly of +TIP comets.

(H) Quantitative analysis of the ratio of the comet-to-cytosolic mCherry fluorescence intensities in cells expressing OptoTIP-SAW under dark (black) and illuminated (blue) conditions (built-in 488-nm excitation channel, 1% output), demonstrating efficient, light-dependent +TIP severing. n = 16 cells from three independent biological replicates.

Scale bars, 5 μm. Error bars denote SEM.

Video S4. Reversible light-induced redistribution of OptoMotor in HeLa cells, related to Figure 4

mCh-tagged OptoMotor showed peripheral accumulation upon external blue light illumination (470 nm, 40 μW/mm²) and cytosolic redistribution after light withdrawal in HeLa cells, demonstrating reversible, light-dependent control of subcellular localization.

Because the C-terminal tail domain of kinesin-1 governs cargo specificity and lysosomal transport,^53,54 we next examined whether OptoMotor can be coupled with the naturally evolved tail domain to drive cargo movement and functional signaling in a light-dependent manner. Appending the native tail domain (residues 907–1027) to OptoMotor yielded OptoMotor-Tail (Figure 4J). In HeLa cells co-expressing OptoMotor-Tail, the majority of LAMP1-GFP-labeled lysosomes accumulated near the nucleus with minimal bidirectional movement in the dark, whereas blue light stimulation triggered their outward movement toward the cell periphery (Figure 4K). This light-driven redistribution demonstrates that OptoMotor-Tail restores cargo recognition and motor-driven lysosomal transport in a controllable manner.

Peripheral redistribution of lysosomes promotes mTORC1 activation by positioning the organelles closer to Rheb-enriched membranes and growth factor-responsive signaling domains at the cell periphery.^{55,56,57,58,59} Consistent with this, photoactivation of OptoMotor-Tail and the ensuing peripheral repositioning of lysosomes resulted in a marked increase in phosphorylation of ribosomal protein S6 kinase (S6K) at T389 (Figure 4L), a hallmark readout of mTORC1 activation that reflects enhanced anabolic signaling and protein synthesis.^60,61 Thus, light-induced recruitment and activation of OptoMotor-Tail directly promote lysosomal repositioning, which in turn boosts mTORC1 signaling through spatial coupling of the lysosomal nutrient-sensing machinery.

Together, these results establish OptoMotor as an optogenetic actuator that reconstitutes intracellular transport and functionally couples cytoskeletal dynamics to signaling output, providing compelling evidence that spatial control of lysosomal trafficking can causally modulate mTORC1 activity.

Optogenetic engineering of spastin as a modular MT-severing tool

Spastin, a hexameric AAA ATPase, catalyzes MT severing by extracting tubulin subunits in an ATP-dependent manner.⁶² Its activity is orchestrated by multiple domains, including an N-terminal MIT domain that governs subcellular targeting, a MT-binding domain (MTBD) that recognizes tubulin C-terminal tails, and a C-terminal AAA domain that mediates ATP hydrolysis and filament severing.^63,64 Because enzymatic activation requires hexamerization, we reasoned that light-induced oligomerization could be leveraged to reconstitute spastin activity in a reversible and tunable fashion.

Building on the OptoMT and OptoTIP scaffolds, we employed a truncation-guided strategy to develop a light-inducible MT-severing toolkit (Figure 5A). Spastin truncations spanning residues 2–616, 228–616, 245–616, 270–616, and 343–616 were fused to OptoMT and expressed in HeLa cells, and the MT network integrity was assayed with GFP-OligoMT, a genetically encoded reporter we previously developed to visualize the MT cytoskeleton.⁶⁵ Among the tested constructs, spastin_228–616 produced the most pronounced filament fragmentation and fluorescence loss after 2 h of blue light illumination (Figures S6A and S6B). Because spastin_228–616 retains the MTBD region, we speculated that residual MT engagement could confer background activity. To test this independent of OptoMT targeting, we expressed mCherry-CRY2-spastin_228–616 and observed partial MT disruption upon blue light stimulation, evidenced by cytosolic diffusion of the OligoMT signal with a concomitant reduction in filamentous structures (Figure S6C).

To eliminate this basal activity, we introduced charge-neutralized mutations (residues 309–311 KKK→ QQQ, K3/Q3) within the MT-binding domain, which minimized basal severing and preserved MT integrity (Figure S6D). When the catalytically competent but non-MT-binding spastin_228–616-K3/Q3 mutant (designated SAW, which stands for severing activated by wavelength) was appended to either OptoMT or OptoTIP (designated OptoMT-SAW or OptoTIP-SAW, respectively), both chimeras triggered robust and spatially confined MT disassembly upon blue light exposure (Figures S7A–S7E). These results indicate that efficient severing requires both MT targeting and light-driven oligomerization. Upon activation, OptoMT-SAW induced progressive fragmentation and cytoplasmic dispersion of OligoMT-labeled filaments (Figures 5B and 5C), with a severing half-life of approximately 45 min (Figure 5D).

Likewise, OptoTIP-SAW efficiently severed MT plus-end structures, as indicated by the progressive loss of comet-like +TIPs visualized with EB1-GFP (Figure 5E). Under repeated light-dark cycles, blue light selectively recruited OptoTIP-SAW to defined ROIs, inducing localized severing of +TIPs while neighboring non-illuminated cells retained a diffuse cytoplasmic signal. After 20 min in the dark, OptoTIP-SAW remained responsive, relabeling and severing +TIPs exclusively within the previously illuminated ROI (Figure 5F). To quantify severing kinetics while avoiding potential artifacts from EB1-based labeling, we employed OligoTIP, a genetically encoded +TIP marker that does not interact with OptoTIP constructs.⁶⁵ Analysis of comet-to-cytosol fluorescence ratios revealed rapid loss of +TIP integrity, with a severing half-life of around 5.3 min (Figures 5G and 5H; Video S5). These findings establish OptoTIP-SAW as a reversible, spatially programmable optogenetic module for precise disassembly of dynamic MT plus-end structures in living cells. Together, these results establish OptoMT-SAW and OptoTIP-SAW as single-component programmable optogenetic actuators (collectively termed OptoSAW) for targeted MT disassembly in living cells.

Video S5. Light-induced severing of MT plus-end structures by OptoTIP-SAW visualized with the +TIP marker EB1-GFP, related to Figure 5

Time-lapse confocal imaging of HeLa cells co-expressing mCh-OptoTIP-SAW (red) showed the progressive loss of comet-like +TIPs labeled by EB1-GFP (green) upon blue light illumination (built-in 488-nm excitation channel, 1% output).

OptoSAW applied to interrogate cell signaling and organelle transport

We next used OptoSAW to examine how acute disruption of dynamic MTs reshapes STIM1-mediated cell signaling and organelle trafficking in living cells. While STIM1 primarily functions as an endoplasmic reticulum (ER) calcium sensor, its interaction with EB1 couples the ER to dynamic MT plus-ends, promoting membrane remodeling and modulating the timing of its translocation to ER-plasma membrane (PM) junctions, thereby regulating store-operated calcium entry (SOCE) (Figure 6A).^66,67 We first examined how MT disassembly influences calcium influx. Depletion of ER calcium stores using thapsigargin (TG) elicited comparable calcium influx, as assessed by GCaMP6s fluorescence, in both control and OptoTIP-SAW-expressing cells after 30 min of illumination (Figure 6B). However, STIM1 puncta formed more rapidly at ER-PM junctions following MT disassembly (Figures 6C and 6D). This is consistent with the model in which EB1 binding tethers STIM1 to growing MT plus-ends, transiently delaying its translocation to the ER-PM junctions and preventing excessive calcium entry (Figure 6A). Disruption of MT plus-end dynamics therefore accelerates STIM1 recruitment to ER-PM junctions without altering overall SOCE amplitude. Together, these results suggest that while MTs are dispensable for the magnitude of SOCE, they probably fine-tune the temporal dynamics of STIM1 activation.

Figure 6.

OptoTIP-SAW dissects microtubule-dependent signaling and organelle transport in real time

(A) Schematic illustrating that disruption of MT plus-ends releases activated STIM1, promoting its translocation to ER-PM junctions and subsequent ER membrane remodeling.

(B) Representative fluorescence traces (left) show Ca²⁺ influx in control cells (blue, intact MTs) and OptoTIP-SAW-expressing cells (red, MT severed). Bar graph on the right shows the quantification of SOCE amplitude from three independent biological replicates (each dot represents the mean of 30–60 cells). Assays were performed in GCaMP6s-HeLa stable cells expressing mCherry-OptoTIP (non-severed control, blue) or mCherry-OptoTIP-SAW (red).

Cells were illuminated for 30 min (external, 470 nm, pulses of 6 s on and 60 s off, 40 μW/mm²) before store depletion with thapsigargin (1 μM).

(C and D) Accelerated STIM1 activation kinetics following MT disassembly.

(C) Confocal of HeLa cells stably expressing STIM1-mCherry and ORAI1-GFP, either without (blue, intact MTs) or with (red, MT severed) co-expression of miRFP-OptoTIP-SAW. Cells were illuminated for 2 h (external, 470 nm, pulses of 6 s on and 60 s off, 40 μW/mm²) before ER store depletion with 1 μM TG, showing faster formation of STIM1 puncta in the MT-severed condition.

(D) Quantification of STIM1 puncta formation kinetics following TG-induced store depletion.

(E and F) OptoTIP-SAW enables light-controlled disruption of lysosomal transport. (E) Merged confocal images showing the subcellular distribution of LAMP1-GFP-labeled lysosomes at the indicated time points. Top left, merged images comparing 1 min (green) and 3 min (red); top right, 25 min (green) and 28 min (red). Bottom, kymograph along the dashed line (top left) illustrating progressive attenuation of lysosome motility over 28 min. (F) Quantification of motile LAMP1-GFP-labeled lysosomes in cells with or without OptoTIP-SAW expression.

Also see Video S6. Scale bars, 5 μm. Error bars denote SEM.

We next applied OptoSAW to investigate how dynamic MTs contribute to organelle trafficking, focusing on lysosomal transport as a representative example. Bidirectional lysosome transport along MT tracks is essential for cellular adaptation and homeostasis in response to environmental cues and stress.^57,68 For instance, nutrient deprivation drives retrograde lysosome movement toward MT minus-ends near the centrosome via cytoplasmic dyneins, whereas nutrient repletion or growth factor stimulation promotes kinesin-mediated anterograde lysosome movement toward MT plus-ends at the cell periphery.^55,69 We leveraged OptoTIP-SAW to assess how disruption of MT dynamics affects lysosomal trafficking. In control HeLa cells, LAMP1-GFP-labeled lysosomes exhibited canonical bidirectional, stop-and-go motion with dynamic redistribution over time (lower; Video S6). In contrast, co-expression of OptoTIP-SAW caused a progressive decline in lysosomal motility under blue light exposure, culminating in transport arrest within 30 min (Figure 6E and Video S6). Kymograph analysis further revealed a shift from dispersed to stationary trajectories, confirming a time-dependent reduction in motile lysosomes (Figure 6F). Collectively, these results demonstrate that light-induced disassembly of MT plus-ends by OptoTIP-SAW effectively impairs MT-based lysosomal transport, confirming the critical role of dynamic MTs in sustaining organelle motility.

Video S6. Light-induced impairment of lysosomal motility by OptoTIP-SAW, related to Figure 6

Top, Time-lapse confocal imaging of HeLa cells co-expressing LAMP1-GFP (green) and mCh-OptoTIP-SAW (red). Upon blue light illumination (built-in 488-nm excitation channel, 1% output), lysosomal motility decreased progressively, leading to near-complete immobilization after approximately 15 min as a result of light-induced MT disruption.

Discussion

MTs form the dynamic scaffold that supports cellular organization, acting not only as structural elements but also as platforms for force generation, intracellular cargo transport, organelle positioning, and signaling coordination.^70,71,72,73 Their highly adaptable nature, shaped by an intricate network of MAPs, confers remarkable architectural plasticity while enabling rapid reorganization in response to cellular cues. Yet this very complexity has made it challenging to define how local MT dynamics translate into functional outcomes within living cells. Conventional approaches for labeling or perturbing MTs, such as pharmacological agents or genetic manipulation, lack the temporal control, reversibility, and subcellular precision needed to capture transient remodeling events. While a growing repertoire of optogenetic tools has been developed to enable spatiotemporal control over specific aspects of MT regulation, such as plus-end dynamics, severing, or motor recruitment, these approaches are generally based on two-component systems and are implemented using distinct engineering strategies.

In this study, we present a single-component, modular optogenetic framework for manipulating diverse MT-associated processes. A central advance of this toolkit lies in its shared engineering strategy, which harnesses light-dependent clustering to program MT engagement. The choice of CRY2 as the central photosensory module was motivated by several practical and mechanistic considerations. Unlike two-component dimerization systems, blue-light-induced CRY2 oligomerization enables single-component designs in which the photosensory domain and MT-regulatory motif are encoded within the same polypeptide, simplifying delivery and maintaining fixed stoichiometry. This feature is advantageous for MT applications, where expression imbalance between two components can alter cytoskeletal localization or generate variable responses.

In summary, this single-component optogenetic platform provides a complementary toolkit that integrates reversible visualization, local biochemical modification, motor regulation, and filament remodeling within a unified photoregulatory framework, which prioritizes modularity, fixed stoichiometry, and ease of recombination. Through programmable control of MT dynamics, it transforms optogenetics from a localization-based perturbation tool into a quantitative system for reconstructing cytoskeletal logic in living cells. The ability to optically build, reshape, and dismantle MT networks in real time establishes a foundation for decoding how cytoskeletal architecture orchestrates intracellular signaling, organelle positioning, and disease-related remodeling.

Limitations of the study

Several technical limitations or considerations warrant mention. Blue light activation limits penetration depth and can introduce phototoxicity, motivating future development of red-shifted or near-infrared photoreceptors to extend in vivo applicability.^74,75,76 CRY2-based oligomerization exhibits slower dark-state recovery than LOV2-based systems, which can limit temporal resolution and spatial confinement under prolonged illumination.²⁴ Accordingly, the tools described here are best suited for applications that benefit from sustained or integrative engagement with the MT network, rather than rapid on-off switching. In addition, the magnitude of CRY2 clustering depends on expression level and temperature, suggesting that standardized promoters or knockin models will enhance reproducibility. Enzymatic modules such as αTAT1 may exceed physiological catalytic rates and thus require calibration to native activity. Sustained activation of OptoSAW can cause irreversible lattice loss, emphasizing the need for optimized illumination duty cycles. Finally, achieving multiplexed control across distinct wavelengths remains an engineering challenge; the addition of spectrally orthogonal switches or self-limiting degradation motifs could refine temporal precision and expand multicolor compatibility.

Resource availability

Lead contact

Requests for further information, resources, and reagents should be directed to and will be fulfilled by the lead contact, Yubin Zhou (yubinzhou@tamu.edu).

Materials availability

This study did not generate new unique reagents. All plasmids generated in this study are available from the lead contact.

Data and code availability

• •All data reported in this paper will be shared by the lead contact upon request.
• •This paper does not report original code.
• •Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Acknowledgments

This work was supported by the National Institutes of Health (R01GM144986 and R35GM163674 to Y.Z.; R35GM144237 to D.J.R.; and R01CA240258, R01DK132286, and R35HL166557 to Y.H.) and the Welch Foundation (A-2310-20260402 to Y.Z.). We thank Shaohe Wang and Karen Oegema for sharing the unpublished C. elegans strain OD2765. Some strains were provided by the Caenorhabditis Genetics Center, which is funded by the NIH Office of Research Infrastructure Programs (P40 OD010440).

Author contributions

Y.Z., Y.H., and G.M. conceived the ideas and directed the work. G.M. and Y.Z. designed the study. G.M., X.L., M.C., D.D., T.-H.L., and Y.Z. designed and generated all the plasmid constructs. G.M., X.L., Y.H., and Y.Z. developed and characterized the optogenetic tools. G.M., T.D., and D.J.R. performed C. elegans studies. G.M., Y.Z., X.L., and T.-H.L. extended the applications of tools. G.M., X.L., T.-H.L., and Y.Z. analyzed the data. G.M., X.L., T.-H.L., and Y.Z. wrote the manuscript, with input from all other co-authors.

Declaration of interests

The authors declare no competing interests.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work, the authors used ChatGPT in order to improve readability of the manuscript. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Mouse monoclonal anti-α Tubulin (DM1A)	Santa Cruz Biotechnology	Cat# sc-32293; RRID: AB_628412
Mouse monoclonal anti-acetylated α Tubulin (6-11B-1)	Santa Cruz Biotechnology	Cat# sc-23950; RRID: AB_628409
Rabbit polyclonal anti-detyrosinated α Tubulin	Abcam	Cat# ab48389; RRID: AB_869990
Mouse monoclonal anti-GAPDH	Thermo Fisher Scientific	Cat# MA5-15738; RRID: AB_10977387
Goat anti-Mouse Alexa Fluor™ 488	Thermo Fisher Scientific	Cat# A-11001; RRID: AB_2534069
Goat anti-Mouse Alexa Fluor™ 555	Thermo Fisher Scientific	Cat# A-21422; RRID: AB_2535844
Goat anti-Rabbit Alexa Fluor™ 488	Thermo Fisher Scientific	Cat# A-11008; RRID: AB_143165
Goat anti-Rabbit Alexa Fluor™ 555	Thermo Fisher Scientific	Cat# A-21429; RRID: AB_2535850
Goat anti-Mouse, HRP	Thermo Fisher Scientific	Cat# 31430; RRID: AB_228307
Goat anti-Rabbit, HRP	Thermo Fisher Scientific	Cat# 31460; RRID: AB_228341
Chemicals, peptides, and recombinant proteins
KOD Hot Start DNA Polymerase	EMD Millipore Corporation	Cat# 71086
NEBuilder HiFi DNA Assembly Master Mix	New England BioLabs	Cat# E2621L
T4 DNA Ligase	New England BioLabs	Cat# M0202T
Restriction endonucleases	New England BioLabs	N/A
QuikChange Multi Site-directed Mutagenesis Kit	Agilent Technologies	Cat# 210514
Lipofectamine 3000	Thermo Fisher Scientific	Cat# L3000015
Nocodazole (CAS, 31430-18-9)	Sigma-Aldrich	Cat# M1404
Thapsigargin (CAS, 67526-95-8)	Sigma-Aldrich	Cat# 586005
Tubulin Tracker™ Deep Red	Thermo Fisher Scientific	Cat# T34077
Triton X-100	Sigma-Aldrich	Cat# T8787
Normal Goat Serum	Thermo Fisher Scientific	Cat# 50062Z
DAPI	Sigma-Aldrich	Cat# D9542
Trypan Blue Solution, 0.4%	Thermo Fisher Scientific	Cat# 15250061
West-Q Pico Dura ECL substrate	GenDEPOT	Cat# W3653-020
Experimental models: Cell lines
HeLa	ATCC	Cat# CRM-CCL-2
HEK293	ATCC	Cat# CRL-1573
SK-MEL-28	ATCC	Cat# HTB-72
C2C12	ATCC	Cat# CRL-1772
U87	ATCC	Cat# HTB-14
NIH3T3	ATCC	Cat# CRL-1658
MIA PaCa-2	ATCC	Cat# CRM-CRL-1420
COS-7	ATCC	Cat# CRL-1651
MEF	ATCC	Cat# CRL-2991
Neuro 2A	ATCC	Cat# CCL-131
H9C2	ATCC	Cat# CRL-1446
A549	ATCC	Cat# CRM-CCL-185
Experimental models: Organisms/strains
C. elegans: Strain DV2162: dpy-20(e1362cs) IV 5 × outcrossed	Caenorhabditis Genetics Center	DV2162
C. elegans: Strain DV3482: itSi916[pOD2050/pSW384; dpy-7p > GFP::tbb-2 3'UTR + Cbr-unc-119(+)] I; dpy-20(e1362cs) IV	Caenorhabditis Genetics Center	DV3482
C. elegans: Strain DV3515: itSi916[pOD2050/pSW384; dpy-7p>GFP::tbb-2 3'UTR + Cbr-unc-119(+)] I; dpy-20(e1362cs) IV; reEx210[pTD61[dpy-7p>mCherry::OptoTip] + pMH86(dpy-20(+))]	Caenorhabditis Genetics Center	DV3515
C. elegans: Strain DV3516: itSi916[pOD2050/pSW384; dpy-7p>GFP::tbb-2 3'UTR + Cbr-unc-119(+)] I; dpy-20(e1362cs) IV; reEx211[pTD61[dpy-7p>mCherry::OptoMT] + pMH86(dpy-20(+))]	Caenorhabditis Genetics Center	DV3516
C. elegans: Strain OD2765: itSi916[pOD2050/pSW384; dpy-7p>GFP::tbb-2; Cbr-unc-119(+)] I; unc-119(ed3) III	Unpublished, gift from Dr. Shaohe Wang and Dr. Karen Oegema	OD2765
Recombinant DNA
mCh-OptoMT	This paper	N/A
mCh-OptoTIP	This paper	N/A
mCh-OptoMotor	This paper	N/A
mCh-Opto-αTAT1	This paper	N/A
mCh-OptoMT-SAW	This paper	N/A
mCh-OptoTIP-SAW	This paper	N/A
Software and algorithms
NIS-Elements imaging software (version 4.5 1.00)	Nikon Instruments Inc.	https://www.microscope.healthcare.nikon.com/products/software/nis-elements/software-resources
GraphPad Prism 10.5.0	GraphPad Software	https://www.graphpad.com
FlowJo v10	BD Biosciences	https://www.flowjo.com
ImageJ (version 2.16.0/1.54p).	NIH	https://imagej.net/ij/download.html
Other
4-Chamber 35mm glass bottom dish	Cellvis	Cat# D35C4-20-1.5-N
Nikon Eclipse Ti-E microscope equipped with the A1R-A1 confocal module	Nikon Instruments Inc.	N/A
470 nm Blue LED Array Light Source	ThorLabs Inc.	Cat# LIU470A
DeltaVision imaging workstation	GE Healthcare	N/A
ChemiDoc Imaging System	Bio-Rad	N/A
BD LSRII flow cytometer	BD Biosciences	N/A

Experimental model and study participant details

Cell lines

HeLa and other cell lines, including HEK293, SK-MEL-28, C2C12, U87, NIH3T3, MIA PaCa-2, COS-7, MEF, Neuro 2A, H9C2, and A549, were obtained from ATCC and maintained at 37°C with 5% CO2 in complete cell-culture medium according to recommendations from the supplier. DNA transfection was carried out by using Lipofectamine 3000 (Life Technologies) following the manufacturer’s instructions. For live-cell or fixed-cell imaging experiments, cells were seeded in four-chamber 35-mm glass-bottom dishes (D35C4-20-1.5-N, Cellvis) at 20–40% confluency one day before transfection.

C. elegans strains and handling

All strains were derived from the N2 Bristol wild type (see strain information in Table S1). Animals were cultured at 20°C on 5 cm NG agar plates with OP50 E. coli as food.⁷⁷ Standard C. elegans nomenclature was followed.⁷⁸ To provide a non-fluorescent selectable marker for transgenesis, the mutation dpy-20(e1362) was crossed into the strain OD2765, which harbors the itSi916 single-copy transgenic insertion expressing GFP::TBB-2.⁷⁹ Rescuing dpy-20(+) plasmid pMH86 was co-injected with the respective plasmid encoding mCherry-OptoMT (dpy-7 p >mCherry::OptoMT) or mCherry-OptoTIP (dpy-7 p >mCherry::OptoTIP) to generate transgenic extrachromosomal arrays; non-Dpy progreny co-expressed mCherry:OptoMT or mCherry:OptoTIP in epithelial cells expressing GFP::TBB-2 (β-tubulin).

Method details

Plasmid construction

The plasmid templates for EB1 (#17234), CLIP170 (#54044), CAMSAP1 (# 59036),CAMSAP2 (#59037), KIF5A (#166954), and spastin (#134461) were purchased from Addgene. To generate OptoMT, we firstly amplified the PHR domain (residues 1–498) of Arabidopsis thaliana CRY2 (Addgene, #70159) by standard PCR and then inserted the fragment into modified pmCherry-C1 and pEGFP-C1 vectors (Clontech), followed by the insertion of multiple MT binding domains, derived from EB1, CLIP170, KIF5A, or CAMSAP1 at the BspEI and EcoRI/BamHI sties, as well as a 3×(SGGGGG) flexible linker between CRY2 and the MT binding domain. For OptoTIP, the EB1-binding SxIP motifs were either directly synthesized, derived from DST (residues 5469–5485), DST (residues 5474–5485), or APC (residues 2786–2824), by Integrated DNA Technologies or amplified via standard PCR. These fragments were inserted to replace the MT binding domain within OptoMT to generated mCh-OptoTIP or EGFP-OptoTIP. OptoTIP variants were subsequently made using the QuikChange multi site-directed mutagenesis Kit (Agilent). To produce LOV2-based OptoTIP, LOV2 from oat phototropin was amplified and then inserted between CRY2 and the SxIP motif. The linker and LOV2 junction regions were further optimized by standard PCR. Opto-αTAT1 (αTAT1-mCh-CIBN-NLS-LEXY) was constructed by using the HiFi DNA assembly method. αTAT1, mCherry, CIBN and LEXY (LOV2-NES) were amplified by standard PCR from pEF5B-FRT-GFP-αTAT1 (Addgene, #27099), pmCherry-C1, CIB1-CreC(N1) (Addgene, #75367) and NLS-mCherry-LEXY (Addgene, #72655), respectively. The amplified fragments were ligated by using the NEBuilder HiFi DNA assembly enzyme (New England Biolabs).

To generate OptoMotor constructs, the sequences of motor-contained domains from kinesin-1 (mouse KIF5A, Addgene, #166954 and mouse KIF5B, Addgene, #31604), kinesin-3 (mouse KIF1B, DNAsu, #MmCD00083530 and human KIF1C, DNAsu, #HsCD00438806), and kinesin-4 (human KIF4B, DNAsu, #HsCD00295076) were amplified by PCR and inserted upstream of CRY2 in either the pEGFP-N1 or pmCherry-N1 vector. To generate OptoMotor constructs with cargo-binding specificity, the C-terminal tail of KIF5A (residues 803–1027) was PCR-amplified and inserted downstream of the CRY2 domain using BspEI and BamHI sites. Organelle-specific targeting was achieved by co-expressing CIBN fused to mitochondrial (AKAP1) or lysosomal (LAMP1) membrane anchors.

To construct the OptoSAW constructs, truncated MT-severing domains of human spastin (Addgene, #134461) were PCR-amplified and fused to the C-terminus of OptoMT or OptoTIP. To minimize background MT binding of spastin, a triple-lysine motif (K310-K312) critical for MT association was mutated to glutamines (KKK-QQQ) using the QuikChange multi site-directed mutagenesis kit (Agilent). All plasmids were verified by Sanger sequencing prior to downstream applications.

Immunostaining

HeLa Cells were seeded on four-chamber 35-mm glass-bottom dishes and cultured until reaching 60–80% confluency. Cells were then transfected with the indicated plasmids using Lipofectamine 3000 following the manufacturer’s instructions. 16–24 h post-transfection, cells were either kept in the dark or exposed to blue light illumination (470 nm, 40 μW/mm²) with the indicated pulse and duration. The cells were fixed with 4% paraformaldehyde in PBS for 20 min at room temperature (RT), rinsed three times with PBS, and permeabilized with 0.1% Triton X-100 in PBS for 10 min at RT. Fixed cells were then blocked with 10% goat serum (Thermo Fisher, #50062Z) for 1 h at RT and incubated with a mouse anti-α-tubulin antibody (1:200 dilution) overnight at 4°C. After three washes with PBST, cells were incubated with Alexa Fluor 488-conjugated IgG secondary antibody (1:500 dilution) for 1 h at RT. Nuclei were counterstained with DAPI, and samples were washed thoroughly and stored in PBS prior to imaging. Confocal images were acquired immediately using a Nikon A1R microscope equipped with a 60× oil-immersion objective. For immunostaining of acetylated α-tubulin and detyrosinated tubulin, the same protocol was followed except that cells were exposed to pulsed blue light illumination (470 nm, 40 μW/mm², 3 s on and 30 s off) for 0, 10, 30, or 60 min for light-induced α-tubulin acetylation or 0, 20, 60, or 90 min for light-induced tubulin detyrosination prior to fixation.

Confocal imaging and image analysis

Fluorescence imaging was mainly performed on a Nikon Eclipse Ti-E microscope equipped with an A1R-A1 confocal module with LU-N4 laser sources (argon-ion: 405 and 488 nm; diode: 561 nm), CFI (chrome-free infinity) plan Apochromat VC series objective lenses (60× oil or 40 × oil), and a live cell culture cage to main the temperature at 37°C with 5% CO₂. Photoactivation with persistent illumination or pulses of repeated dark-light cycles was achieved by an external blue light source (470 nm, 40 μW/mm², ThorLabs Inc., Newton, NJ, USA) or by utilizing the built-in 488-nm excitation channel (5% input or at a power density of 10 μW/mm²). In some experiments, images were acquired using a a DeltaVision imaging workstation (GE Healthcare) equipped with a 100 ×/1.45 oil lens and a CoolSNAP EMCCD camera to achieve higher temporal resolution.

All the acquired images were analyzed using the NIS-Elements imaging software (version 4.5 1.00), and the results were plotted using the Prism 10.5.0 (673) software (GraphPad). Cytosolic fluorescence intensities were quantified using the semi- or fully-automatic image analysis tool in the NIS-Elements software package. Regions of interest (ROIs), such as MT, MT plus-ends (comets), and adjacent cytosolic areas, were defined and measured using the “Intensity Line Profile” tool by drawing a line to extract fluorescence distribution profiles. The fluorescence intensity of the target protein within each ROI was quantified for further statistical analysis. Typically, 6–8 regions per cell were analyzed to obtain averaged values.

Cell cycle and viability analysis

24 h post-transfection, HeLa cells expressing either mCh-OptoMT or mCh-CRY2 (as control) were washed with PBS and fixed in ice-cold 70% ethanol at 4°C for 30 min. After washing with PBS, the cells were incubated with 1 mg/mL of DAPI (Sigma D9542) and analyzed by flow cytometry. Histograms of cell cycle distribution were acquired using a BD LSRII flow cytometer (BD Biosciences), and the proportions of cells in G₀/G₁, S, and G₂/M phases were determined using FlowJo v10 software. Each sample was assayed in triplicate. Cell viability was assessed by the standard trypan blue staining assay as described previously.⁸⁰

Western blot analysis

On day 1, cells were trypsinized and washed three times with ice-cold PBS, then lysed directly in RIPA buffer supplemented with 1× protease inhibitor cocktail and phosphatase inhibitor cocktail for 30 min on ice. Lysates were centrifuged, and the resulting supernatant was transferred to new tubes and denatured at 95°C for 5 min in 1× SDS loading buffer (100 mM Tris-HCl, 4% SDS, 0.2% bromophenol blue, 20% glycerol, 200 mM DTT, pH 7.4). Equal amounts of protein were resolved on a 10% SDS-PAGE gel and transferred onto nitrocellulose membranes. Membranes were incubated with the indicated primary antibodies overnight at 4°C. On day 2, the membranes were washed and incubated with secondary antibodies for 1 h at RT and visualized using the ChemiDoc Imaging System (Bio-Rad) with West-Q Pico Dura ECL substrate (GenDEPOT). Densitometric analysis of immunoblot bands was performed using the Gel Analysis function in ImageJ (NIH; version 2.16.0/1.54p). Band intensities were quantified after background subtraction and normalized to GAPDH or the indicated reference proteins.

Real-time intracellular Ca²⁺ measurements

Ca²⁺ influx was monitored in HeLa cells stably expressing the green Ca²⁺ indicator GCaMP6s and transiently transfected with either mCherry-OptoTIP or mCh-OptoTIP-SAW. Time-lapse fluorescence imaging was performed with 10 s intervals, and images were analyzed using NIS-Elements AR software (version 4.5 1.00) with manually defined regions of interest. To measure store-operated Ca²⁺ entry (SOCE), cells were incubated in Ca²⁺ free HBSS buffer (107 mM NaCl, 7.2 mM KCl, 1.2 mM MgCl₂, 11.5 mM glucose and 20 mM HEPES-NaOH, pH 7.2) 5 min prior to imaging. ER Ca²⁺ stores were depleted by using 1 μM thapsigargin (TG), followed by addition of CaCl₂ to a final concentration of 2 mM to induce Ca²⁺ influx.

Quantification and statistical analysis

All data are presented as mean ± sem unless otherwise noted. Sample sizes (n) were listed for each experiment. For Opto-αTAT1 and SxIP tag-related experiments, two-tailed Student’s t test was used to analyze significant differences between different groups. For blue light-induced dimerization and oligomerization, two-tailed Student’s t test was used to analyze significant differences. For all statistics, ns, p ≥ 0.05; ∗∗∗p < 0.001.

Published: July 20, 2026