Kinesin-1-transported liposomes prefer to go straight in 3D microtubule intersections by a mechanism shared by other molecular motors

Abstract

Kinesin-1 ensembles maneuver vesicular cargoes through the three-dimensional (3D) intracellular microtubule (MT) network. To define how such cargoes navigate MT intersections, we first determined how many kinesins from an ensemble on a lipid-based cargo simultaneously engage a MT, and then determined the directional outcomes (straight, turn, terminate) for liposome cargoes at perpendicular MT intersections. Run lengths of 350-nm diameter liposomes decorated with up to 20, constitutively active, truncated kinesin-1 KIF5B (K543) were longer than single motor transported cargo, suggesting multiple motor engagement. However, detachment forces of lipid-coated beads with ~20 kinesins, measured using an optical trap, showed no more than three simultaneously engaged motors, with a single engaged kinesin predominating, indicating anticooperative MT binding. At two-dimensional (2D) and 3D in vitro MT intersections, liposomes frequently paused (~2 s), suggesting kinesins simultaneously bind both MTs and engage in a tug-of-war. Liposomes showed no directional outcome bias in 2D (1.1 straight:turn ratio) but preferentially went straight (1.8 straight:turn ratio) in 3D intersections. To explain these data, we developed a mathematical model of liposome transport incorporating the known mechanochemistry of kinesins, which diffuse on the liposome surface, and have stiff tails in both compression and extension that impact how motors engage the intersecting MTs. Our model predicts the ~3 engaged motor limit observed in the optical trap and the bias toward going straight in 3D intersections. The striking similarity of these results to our previous study of liposome transport by myosin Va suggests a "universal" mechanism by which cargoes navigate 3D intersections.

Keywords: cytoskeleton; intracellular transport; mathematical modeling; optical trap; super-resolution imaging.

Fig. 1.

3D MT intersection model system. (A) Schematic representation of 3D MT intersection assay in B, as viewed from above. Beads (gray circles) serve as pedestals from which MTs (green) are suspended. (B) 3D STORM image of suspended MT intersections. Opaque white circles highlight pedestal beads. The color bar represents Z-axis scale. (Scale bar, 3 μm.) (C) Illustration of a liposome cargo (yellow sphere) transported by multiple kinesin-1 motors (purple) approaching an intersection between two MTs (green). Dashed lines indicate the plane which contains the MT intersection. Dashed arrows represent diffusion of kinesin-1 on the liposome surface. (D) Movie frames of a liposome (yellow) proceeding straight through a MT intersection. Outcome shown by dashed green arrow. MT gap = 250 nm. (Scale bar, 500 nm.) Color scale is the same as in B. (E) Movie frames of a liposome (yellow) turning in a MT intersection. Outcome shown by dashed green arrow. MT gap = 130 nm. (Scale bar, 500 nm.) Color scale is the same as in B.

Fig. 2.

Multiple Kinesins can simultaneously engage the MT. (A) Kymographs of cargo motility along a single MT in 2D. Left kymograph is liposome motility with 10 kinesin-1s, right kymograph is single kinesin-1 Qdot motility. The vertical axis represents time and horizontal axis represents distance. Bounding box (yellow) indicates a single run by a kinesin-1-transported Qdot where the horizontal axis gives the run length, the vertical axis gives the event lifetime, where the average velocity is run length divided by event lifetime. (B) Beeswarm plot of cargo run length from experimental (Experiment) and simulated (Model) kinesin-1 motility as a function of motors per cargo. Median with interquartile range shown. Experimental N = 297, 161, and 137 events for 1, 5, and 10 motors per cargo, respectively. Simulated trajectories are cutoff after 15 s, which gives a maximum run length of ~11 μm. Beeswarm plot for MT length distribution (in vitro MTs) as measured in our in vitro motility assays. Significance: ns = P > 0.05, *P < 0.05, ***P < 0.0005. Explicit P-values are shown in SI Appendix, Table S1. (C) Beeswarm plot of cargo velocity from experimental (Experiment) and simulated (Model) kinesin-1 motility as a function of number of motors per cargo. Median with interquartile range shown. Experimental N = 297, 161, and 137 events for 1, 5, and 10 motors per cargo, respectively. Significance: ns = P > 0.05, *P < 0.05, ***P < 0.0005. Explicit P-values are shown in SI Appendix, Table S2. (D) Schematic of optical trap ramp force assay where a trapped lipid-coated bead (yellow) is pulled by a kinesin-1 motor out of the trap (red triangles) center with the trap resistive load proportional to the bead displacement (ΔX). (E) An example optical trap force ramp for a lipid-coated bead transported by multiple kinesin-1 molecules. Inset, example force ramp for a lipid-coated bead transported by a single kinesin-1 molecule. In both traces, raw data are shown in gray while median-filtered data (11-point filter width) are overlaid in black. The displacement from trap center (Right y-axis) is converted to resistive force (Left y-axis) based on the trap stiffness. (F) Histogram of maximum force prior to detachment for beads incubated with 20-fold molar excess of kinesin-1 (gray bins) and single kinesin-1 transported beads (white bins). The green solid curve shows triple Gaussian fit to the multimotor data. N = 213 events from five independent bead preparations. The gold curve shows Gaussian fit of maximum force prior to detachment for single kinesin-1 transported bead. N = 34 events from three independent bead preparations. Inset, histogram of modeled maximum force prior to detachment for kinesin-1 ensembles (gray bins). The dashed green curve shows triple Gaussian fit to the data. N = 202 simulated force ramps. The dashed gold curve shows Gaussian fit to simulated single motor detachment force histogram (white bins). N = 100 simulated force ramps. All fit parameters are shown in Table 1. (G) Cumulative distribution plot of force ramp event lifetimes for experimental (solid) and simulated (dashed) force ramps. In both cases, single motor lifetimes are shown in gold and 20-motor lifetimes are shown in green.

Fig. 3.

Directional outcomes for kinesin–liposome complexes in 2D MT intersections depend on the approach MT and motor number. (A) Schematic of 2D MT intersection assay. Differentially labeled MTs are sequentially affixed to a coverslip with dim MTs first (bottom MT) and bright MTs second (top MT). Events are split up based on whether the liposome approach from the bottom or top MT. Bottom-approach event is shown in the schematic. (B) Timestamp images of a 10-kinesin–liposome complex passing straight (upper sequence) or turning (lower sequence) in a 2D MT intersection. In both cases, the liposome approaches the MT intersection on the dimly fluorescent, bottom MT (Methods). The yellow arrowhead indicates starting liposome position. The white dashed line shows the liposome path. (Scale bar, 500 nm.) (C) Bar graph of directional outcomes for liposomes approaching the intersection on the bottom MT. N = 50, 38, and 60 events for 5, 10, and 20 motors per liposome, respectively. Data from at least three separate liposome preparations. (D) Bar graph of directional outcomes for liposomes approaching the intersection on the top MT. N = 44, 42, and 41 events for 5, 10, and 20 motors per liposome, respectively. Data from at least three separate liposome preparations. (E) Cumulative distribution plot of pauses in 2D intersections for cargoes approaching from the bottom MT. Color indicates number of motors per liposome. (F) Cumulative distribution plot of pauses in 2D intersections for cargoes approaching from the top MT. Color indicates number of motors per liposome.

Fig. 4.

Liposomes can take helical paths on suspended MTs. (A) Schematic illustration of three liposome transport scenarios along a suspended MT tightrope. The liposome can move along the MT in a straight trajectory (magenta), in a helical trajectory (cyan) around the MT or in a random trajectory (black, dashed) with no defined path toward the MT plus-end. (B) Top, example of observed helical liposome trajectory. The green line represents MT, red dots represent liposome position along the MT, and the black curve represents 3D sinusoidal fit (Methods and SI Appendix, Fig. S4). Bottom, example of observed straight liposome trajectory. Horizontal scale bar refers to position along the MT, vertical scale bar refers to liposome center vertical position relative to MT center. (C) Histogram of reciprocal pitches measured in helical liposome trajectories. The black curve is the kernel density function and is for visualization purposes only. N = 23 total events, 15 helical events.

Fig. 5.

Kinesin–liposome complexes preferentially go straight in 3D MT intersections. (A) To-scale schematic of a kinesin–liposome complex (yellow sphere) approaching (moving into the image plane) a 3D MT intersection. Two geometric parameters describe the spatial relationship between the liposome and the intersection, the Gap (d) between the MTs along the Z-axis and the Approach Angle (α), which describes the position of the liposome relative the crossing MT where 0° is up toward and 180° is down away from the crossing MT. (B) Left: Polar plot of the spatial relations (i.e., geometries) between the liposome and the intersecting MTs in terms of d and α. Magenta line divides interaction and noninteraction geometries for liposome with intersecting MT. Right Top: To-scale schematic of an interaction geometry, precise parameters of interaction shown with a blue triangle on the polar plot. Right Bottom: To-scale schematic of a noninteraction geometry, precise parameters of interaction shown with a red triangle on the polar plot. (C) Bar chart of directional outcomes in 3D MT intersections for liposome–MT interaction events only, experimental (solid, N = 100 events) and modeled (crosshatch, N = 178 simulations). (D) Heatmap of modeled (left half) and experimental (right half) straight to turn ratios as a function of approach angle (α) and gap (d). Events are split into 4 zones based on MT gap (>175 nm or <175 nm, i.e., the liposome radius) and approach angle (>75° or <75°), with the fifth zone being noninteraction events, below the magenta border. Each zone is color-coded (see color bar) by the absolute straight-to-turn ratio in that zone (white decimal numbers). Total N = 118 events for experimental dataset, 100 interactions and 18 noninteractions. Number of interactions for a given zone shown as numbers in parentheses. For modeled events, N = 210 events total, 178 interactions and 32 noninteractions. (E) Cumulative distribution plot of pause lifetimes in 3D intersections for experiments color-coded by outcome (green, blue) with the overall pause lifetime in black. (F) Cumulative distribution plot of pause lifetimes in modeled 3D intersections color-coded by outcome (green, blue) with the overall pause lifetime in black.

Fig. 6.

Model-simulated liposome proceeds straight through a MT intersection following a tug of war between motors engaged with each of the intersecting MTs. In all frames: liposome (yellow sphere), kinesin (two heads with a purple tail connected to liposome with a red line representing the magnitude and direction of force on the corresponding kinesin), MTs (green rods with plus-end indicated). (A) Liposome approaches (α = 0°) an intersection (d = 75 nm). (B) A new kinesin engages the crossing MT, entering into a tug of war with the kinesins engaged with the starting MT. (C) The force of the kinesin attached to the crossing MT rotates the liposome around the starting MT and thus underneath the crossing MT. (D) The kinesin on the crossing MT detaches, allowing the liposome to proceed straight through the intersection.