Kinesin-1 activity recorded in living cells with a precipitating dye

Microtubules (MTs) are dynamic polymers that organize cellular architecture and transport, emanating primarily from centrosomes, with the Golgi apparatus serving as a major non-centrosomal MT organizing center. Kinesin-1 is an ATPase motor that transports cargo toward MT plus-ends and is involved in Golgi–ER and ER–Golgi trafficking. In cells, kinesin-1 is largely autoinhibited and becomes active upon cargo binding; only a fraction (~30%) is motile at any time, complicating visualization using tagged motors due to high background from immotile pools. Existing methods rely on fixation, antibody staining, quantum dots, or overexpressed fluorescent fusions, which do not specifically report on motile activity. Quinazolinone-based precipitating dyes (QPDs) can report enzymatic activity via fluorogenic uncaging. The authors sought to develop a QPD derivative (QPD-OTf) that would act as an activity-based substrate to visualize native kinesin-1 activity on MTs in living cells without genetic manipulation, thereby enabling mapping of transport-active MT subsets and motor paths.

The study builds on prior knowledge that: (1) the Golgi is a microtubule-organizing organelle; (2) kinesin-1 preferentially uses stable, modified MTs (acetylated/detyrosinated) and its transport can be perturbed by Taxol; (3) only ~30% of cellular kinesin-1 is active, limiting GFP-tagged motor imaging; and (4) QPD-based precipitating fluorogenic probes have successfully reported hydrolytic enzyme activities (phosphatases, proteases) and reactive species but had not been applied to motor proteins. Previous live-cell cytoskeleton probes (e.g., SiR-Tubulin) and strategies to track motors required perturbations or did not distinguish motile fractions. This work addresses these gaps by introducing an activity-based fluorogenic substrate for kinesin-1.

• Probe design and rationale: QPD-OTf synthesized as a triflate-caged quinazolinone precipitating dye. The triflate renders the molecule water-soluble and non-fluorescent; removal yields insoluble, fluorescent QPD that precipitates at the site of activity.
• Cell lines and culture: U2OS, HeLa, HEK293T, MCF-7, RAW264.7, PTK2 (GFP-Tubulin stable line), and HeLa GFP-Tubulin CRISPR knock-in cells maintained under standard conditions.
• Cellular crystallization assay: Cells incubated with 20 µM QPD-OTf in additive-free DMEM for 20 min to 4 h at 37 °C, 5% CO₂. Imaging by widefield, confocal, and super-resolution (Leica SP8 LIGHTNING) microscopy.
• Cytoskeletal colocalization: Fixed-cell α-tubulin immunostaining and live-cell imaging (GFP-tubulin or SiR-Tubulin) to assess crystal alignment with MTs; phalloidin for F-actin.
• MT perturbations: (i) Stabilization with Taxol (1 µM, 1 h) followed by QPD-OTf (4 h); (ii) Depolymerization by cold treatment on ice (1 h) with subsequent QPD-OTf incubation on ice (4 h). Crystal counts and morphology quantified.
• Golgi localization: Transient expression of mCherry-Giantin to label Golgi; live-cell confocal imaging to assess crystal nucleation at Golgi. Golgi disruption with Brefeldin A (20 µM) prior to and during QPD-OTf incubation to evaluate effects on crystal origin and fiber thickness.
• Kinesin-1 modulation in cells: Transient transfection with truncated kinesin-1 constructs: Kin330-GFP (dominant-negative, non-motile) and Kin560-GFP (constitutively active). Quantified crystal numbers versus controls. Kinesin-1 knockdown (siRNA against KIF5B) with validation by western blot; quantified crystal fluorescence intensity.
• Small-molecule activation: Kinesore (100 µM) treatment ± QPD-OTf in Ringer’s buffer to activate kinesin-1 and assess resultant QPD signal patterns.
• In vitro reconstitution: Precipitation assays in BRB80 buffer with combinations of 20 µM QPD-OTf, 150 nM kinesin-1, 14 µM tubulin/MTs, 1 mM GTP, 2.7 mM ATP, or 2.7 mM AMP-PNP. Fluorescence under 366 nm, confocal imaging of filamentous precipitates. Tested Taxol-stabilized and dynamic MTs without motors as controls.
• FIB-SEM: HeLa cells treated with QPD-OTf (20 µM, 4 h), fixed and embedded; 3D ultrastructural imaging (Zeiss Crossbeam 540), segmentation (Ilastik), rendering (Imaris) to analyze crystal morphology and nucleation centers.
• Molecular docking: Autodock Vina docking of QPD-OTf into kinesin-1 (PDB 3J8Y) ATP-binding pocket and Eg5 (PDB 4AP0) allosteric and nucleotide sites; comparison with ATP/ispinesib poses and binding energies; interpretation of triflate orientation relative to hydrolysis site.
• Statistics: Replicates across microscopy experiments; t-tests applied; sample sizes as indicated in figures; source data provided.
• Data/code availability: Data and code deposited in Zenodo repositories; details provided by authors.

• QPD-OTf generates bright, aster-like fluorescent crystals in live cells within minutes to hours, forming filaments that can deform membranes; crystals dissolve after washout and recovery.
• Crystals co-localize with a subset of microtubules, but not with actin. Super-resolution/live imaging shows alignment of crystal fibers along MT bundles.
• MT integrity and dynamics are required: Taxol stabilization reduced crystal formation by ~75% (p < 0.0001) and yielded thinner fibers; cold-induced depolymerization eliminated crystals.
• Crystals nucleate at the Golgi apparatus: mCherry-Giantin colocalization shows crystal origins at Golgi with radial extension. Brefeldin A–induced Golgi fragmentation produced thinner fibers (thickness reduced by 58%; p = 0.0001) and multiple nucleation foci that still co-localized with Golgi remnants.
• Purified MTs alone do not produce crystals in vitro, indicating need for an additional activity.
• Kinesin-1 dependency in cells: Dominant-negative Kin330 reduced crystal numbers by 87% versus control (p = 0.0064); Kin560 overexpression correlated with crystal-bearing MTs but did not significantly increase crystal counts (Control vs Kin560 n.s., p = 0.065). siRNA knockdown of KIF5B significantly reduced crystal intensity (p = 0.0007) with reduced kinesin-1 confirmed by western blot.
• Kinesore activation disrupted crystal formation and produced diffuse QPD fluorescence, consistent with uncoupled, overactive kinesin-1 motion and supporting dependence on regulated kinesin-1 transport.
• In vitro, kinesin-1 plus MTs are sufficient to produce fluorescent, filamentous QPD precipitates. The strongest signal occurred with kinesin-1 + tubulin/MTs + GTP (sample 4; intensity 12571 arbitrary gray units), while AMP-PNP reduced precipitate (e.g., sample 3 intensity 7027) and ATP modestly reduced it (sample 5 intensity 9573), consistent with competition at the ATP-binding site and requirement of motor activity.
• Docking supports QPD-OTf as a substrate analog binding in the kinesin-1 nucleotide pocket (predicted ΔG ≈ -8.3 kcal/mol) with triflate overlapping the γ-phosphate of ATP, rationalizing motor-dependent triflate cleavage. Docking and cell data argue against Eg5 as the enzyme: mitotic spindle lacked QPD precipitation and Eg5 docking placed triflate away from hydrolysis or toward solvent.
• FIB-SEM reveals crystals with rotational symmetry and hexagonal cross-sections (100–700 nm) and clear nucleation centers, consistent with biological templating along MTs from the Golgi.

The findings demonstrate a first-in-class activity-based fluorogenic substrate for a motor protein that enables direct visualization of native kinesin-1 transport paths in living cells without genetic tagging or fixation. QPD-OTf conversion to an insoluble fluorescent precipitate records kinesin-1 trajectories along selected microtubules, revealing preferential use of specific MT subsets and Golgi-originated transport. Dependence on MT dynamics, Golgi integrity, and kinesin-1 activity (genetic perturbations and small-molecule activation) establishes specificity. In vitro reconstitution with kinesin-1 and MTs is sufficient to produce precipitates, and docking supports binding in the ATP site with the triflate mimicking the γ-phosphate position, suggesting a mechanistic basis for selective activation by kinesin-1 over Eg5. This tool allows mapping of transport-active microtubules within complex networks and may facilitate studies of organelle trafficking, motor regulation, and cytoskeletal organization.

This study introduces QPD-OTf, an activity-based, fluorogenic small-molecule substrate that reports kinesin-1 activity in live cells by precipitating a bright dye along the motor’s path on microtubules. Crystals nucleate at the Golgi and extend radially, reflecting native kinesin-1–mediated transport. Genetic and pharmacological perturbations, in vitro reconstitution, and docking collectively support that QPD-OTf acts as an ATP-site substrate analog for kinesin-1. Future work could: (1) optimize probe sensitivity/specificity and reduce cytotoxicity for long-term imaging; (2) adapt the strategy to other motors (e.g., dynein, kinesin family members) and cargos; (3) quantify transport dynamics by correlating crystal growth kinetics with motor stepping; (4) integrate with super-resolution and correlative EM to map transport highways in diverse cell types and disease models; and (5) elucidate the detailed chemical mechanism of triflate hydrolysis by the motor ATPase.

• Potential sample preparation artifacts in FIB-SEM (e.g., crystals appearing to penetrate the nucleus) due to fixation/dehydration.
• Long exposures (overnight) to 10–20 µM QPD-OTf induce cytotoxicity; viability preserved only with shorter incubations and washout.
• Variability in crystal morphology and kinetics across cell lines; crystals form on a subset of MTs, so not all transport events may be captured.
• Mechanism inferred from docking and activity assays; direct biochemical demonstration of triflate hydrolysis chemistry by kinesin-1 ATPase remains to be fully delineated.
• Kinesore treatment yields diffuse fluorescence and inhibits crystal formation, indicating that probe readout depends on regulated kinesin-1 localization and may be perturbed by strong activators.
• MT-stabilizing or -depolymerizing treatments suppress signals, limiting use in contexts where MT dynamics are intentionally altered.