The DNA tethered bead assay to reduce the z-force in the optical trap.

A) (Left) Conventional single-bead trapping assay where a truncated, biotinylated kinesin is attached to an 860-nm-diameter streptavidin-coated bead. (Right) The DNA-tethered bead assay, where full-length KIF5B is attached to a 510-nm-diameter bead via a 2800 bp DNA handle. B) Characteristic force-generating behavior of kinesin against a stationary trap in both a high and low z-force regime. Stalls are marked with black arrows, while premature detachments are marked with red arrows. C) Stall force of kinesin (mean ± s.e.m.) in the high and low z-force regimes. D) The inverse cumulative distribution function (1-CDF) of the peak forces (median ± s.e.m.) of every processive run that exceeded 1.5 pN hindering force in the high (N = 264) and low (N = 51) z-force regimes. E) 1-CDF of motor stall times. Solid curves represent fits to a double exponential decay to calculate the amplitude (A) and mean lifetime (τ; ± s.e.). The weighted averages of the two decays are 0.27 ± 0.02 s and 1.49 ± 0.08 s for high (N = 85) and low (N = 45) z-forces, respectively.

Kinesin exhibits asymmetric slip bond behavior under force.

A) Example traces of kinesin motility under a 2 pN hindering force in high and low z-forces. The trap position (magenta) is updated to remain 100 nm behind the bead position along the microtubule long axis (blue). The trap stiffness is fixed at 0.02 pN nm-1 so that this separation corresponds to 2 pN. B) Run lengths (fit parameter from single-exponential CDFs, ± s.e.), velocities (mean ± s.e.), and detachment rates (defined as the ratio of run length to velocity of the motor) for forces ranging from –4 pN to +4 pN under high (from left to right, N = 60, 69, 65, 50, 167, 100, 115 runs) and low (from left to right, N = 89, 76, 44, 42, 65, 32, 39 runs) z-force conditions. For –10 pN and –6 pN hindering forces, detachment rates were calculated from the run time distributions under high (N = 133, 167) and low (N = 43, 73) z-force regimes (Figure 2_figure supplement 2), as these conditions yielded minimal forward displacement. Negative and positive forces correspond to hindering (minus-ended) and assisting (plus-ended) directions, respectively. Under hindering forces, kinesin runs several-fold longer distances and has a lower detachment rate in the low z-force regime compared to the high z-force regime.

Kinesin motors team up efficiently under low z-force conditions.

A) Schematic of the high and low z-force assays with three K560 motors bound to a DNA chassis. Left: The 3-motor DNA chassis binds directly to a streptavidin-coated bead. Right: The DNA chassis with a single-stranded DNA overhang hybridizes to the overhang on the long DNA handle attached to the bead. B) Mass photometry measurements with a DNA chassis are performed with a 3-fold excess of kinesin. The Gaussian mixture model identifies the mass and percentage of unbound kinesin (blue curves), chassis with one (blue curves), two (green curves), and three (red curves) motors. Mass populations of unbound motor and chassis with one motor bound cannot be distinguished from each other. C) Example trajectories of beads driven by two and three K560 motors in the high and low z-force regimes. These assays were performed with a 10 or 20-fold excess of kinesin for 2- or 3-motor DNA chassis, respectively. D) Forces (mean ± s.d.) produced by single kinesins or kinesins bound to 2- or 3-motor chassis under high (from left to right, N = 85, 66, 158) and low z-force conditions (from left to right, N = 37, 65, 66). The solid line represents a fit to a power function to calculate the scaling factor (c, ± s.e.). E) Dwell times before slips are plotted as 1-CDF. Weighted average time constants (τ, ± s.e.) are extracted from fitting to a double exponential decay.

Kinesin resists microtubule detachment and backward processive movement under low z forces.

A) High z-forces inherent to optical trapping assays result in rapid detachment of kinesin from microtubules, whereas the motor experiences lower z-forces when carrying intracellular cargos and more persistently binds to microtubules under those conditions. B) (Top) The tug-of-war between kinesin and dynein results in pausing or stalling the cargo movement, as well as slow motility interspersed with frequent reversals, because either motor resists detachment or backward stepping under forces generated by its opponent. (Bottom) Coordination between the opposing motors, such as activation of one motor at a time (reciprocal activation) results in fast unidirectional transport in the direction of an active motor while the inactive motor is carried as a passive cargo.

The list of DNA constructs used in this study.

Purification of kinesin and DNA handle with experimental controls.

A) SDS-Page denaturing gel of constructs used. K560-GFP-SNAP was labeled with biotin, and KIF5B-GFP-SNAP was labeled with a BG-functionalized DNA oligo before eluting the motor from the IgG beads during purification. B) The long DNA handle was run on a 0.8% TAE agarose gel. Left: Handle incubated with a 20-fold excess of complementary Cy3 oligo and imaged in the Cy3 channel of a Typhoon FLA 9500 fluorescence imager. Right: The same gel imaged under UV after staining in 2x GelRed for 40 minutes. C) Full-length KIF5B-GFP-SNAP was incubated with a 10-fold excess of complementary Cy5-labeled oligo and run in a motility assay. The assay was performed in the presence of 10 nM MAP7. D) Kymographs of K560-GFP-biotin (top, N = 89) and Qdot 655 streptavidin quantum dots being transported by oligo-labeled KIF5B attached to the long DNA handle (bottom, N = 53). Histograms of the respective velocities are shown to the right. E) Mechanical demonstration of a bead tethered to an axoneme by the long DNA handle. After tethering to an axoneme, the bead is pulled away from the axoneme by an optical trap. The bead moved freely for ∼1 µm and stayed in that position, demonstrating the formation of a long tether between the bead and the axoneme through the DNA handle. F) 1-CDF kinesin run time under unloaded conditions in single-molecule fluorescence imaging assays (N = 160 for K560 and 88 for KIF5B). Solid curves represent a fit to a single exponential decay function to calculate the lifetime of processive runs (τ, ±S.E.). G) CDF of kinesin run length under unloaded conditions in single-molecule fluorescence imaging assays. Solid curves represent a fit to a single exponential decay function to calculate the mean run length (±S.E.). In D, F, and G, KIF5B assays were performed in the presence of 50 nM MAP7.

Example traces for force-feedback controlled trapping of kinesin with or without a DNA handle.

Example traces for force feedback measurements under the high z-force and low z-force conditions in hindering and assisting directions. Scale bars are 1 s.

Raw histograms and fits to force-feedback controlled trapping of kinesin with or without a DNA handle.

A) CDFs of run length at each condition (blue) and the corresponding fit (red). Under the 4 pN hindering condition, a small number of events demonstrated net negative run lengths. A weighted average was used to calculate the run length shown in Fig. 2. From left to right, N = 3, 55, 69, 65, 50, 167, 100, 115 on the top row, and 11, 78, 76, 44, 42, 65, 32, 39 in the bottom row. B) Histograms of velocities for each condition. From left to right, N = 60, 69, 65, 50, 167, 100, 115 in the top row and 89, 76, 44, 42, 65, 32, 39 in the bottom row. C) 1-CDFs of motor attachment time to the microtubule without net positive displacement under stall or superstall forces. Solid red curves represent a fit to a single exponential decay to calculate the detachment rate (From left to right, N = 167, 133, 73, 43).

Purification and stall force of multi-motor chassis.

A) SDS-PAGE denaturing gel to quantify oligo labeling of K560-GFP-SNAP. Left lane: molecular weight marker. Middle lane: K560-GFP-SNAP. Right Lane: K560-GFP-SNAP labeled with DNA oligo. Arrows show that the oligo-labeled motor is discernible from the unlabeled motor on the gel. B) Gel extraction of multi-motor chassis. The left two lanes show 3- and 2-motor chassis for high z-force measurements before gel extraction. The middle two lanes show 3- and 2-motor chassis for high z-force measurements after gel extraction. The right two lanes show 3- and 2-motor chassis for low z-force measurements after gel extraction.

Model for estimating the percentage of DNA chassis bound to two- or three-motors in mass photometry.

Unbound motor is 3-fold excess of the DNA chassis. p is the probability of binding of kinesin to the DNA chassis. The model assumes no cooperativity between the binding sites on the chassis. Motor bound to chassis was calculated from the probability of chassis multiplied by the number of motors bound to the chassis. Mass photometry cannot distinguish an unbound motor and chassis with one motor bound. p-values were calculated from the ratios of the percentages of distinct mass populations detected by mass photometry.

Stall force histograms (mean ± s.e.) of K560, and 2- or 3-motor chassis under high and low z-force conditions.