Search

The goal of this work was to investigate the mechanisms underlying different Eg5 (kinesin-5) inhibitors and to assess their effects on Eg5-mediated stabilization of dynamic microtubules. Antimitotic Eg5 inhibitors were grouped into those that bind Loop-5 and produce an ADP-like state (monastrol, STLC, ispinesib and filanesib) and those that produce a rigor-like state (BRD9876). To understand their contrasting effects on Eg5 mechanics, inhibitors were analyzed in mixed-motor gliding assays consisting of different ratios of kinesin-1 and Eg5 motors, in which Eg5 “braking” dominates motility. Loop-5 inhibitors increased gliding speeds, consistent with the compounds inducing a weak-binding state in Eg5, whereas BRD9876 slowed gliding, consistent with Eg5 being locked in a rigor state. Quantitative critical detachment force analysis (Arpag et al. Biophys J. 2014) indicates that L5 inhibitors enhance Eg5 detachment whereas the rigor inhibitor BRD9876 generates an Eg5 species with decreased susceptibility of detachment. Biochemical and single-molecule assays demonstrated that BRD9876 acts as an ATP- and ADP-competitive inhibitor with a KI of 4 nM. Consistent with its recently reported microtubule polymerase activity (Chen and Hancock, Nature Comm. 2015:8160) and its dominant two-head-bound strong-binding population (Chen et al., JBC 2016:M116), Eg5 was shown to stabilize microtubules against depolymerization in vitro following taxol washout. This stabilization was eliminated in monastrol but was enhanced by BRD9876, consistent with strong binding of Eg5 to microtubules enhancing microtubule stability. Finally, in metaphase-arrested RPE-1 cells, STLC promoted spindle collapse, whereas BRD9876 did not. Thus, different Eg5 inhibitors inhibit spindle formation through contrasting mechanisms and rigor inhibitors are predicted to paradoxically stabilize microtubule arrays in cells.

The kinesin-2 family motor KIF3A/B coordinates with dynein to bidirectionally transport intraflagellar particles, melanosomes and neuronal vesicles. Compared to kinesin-1, kinesin-2 is less processive and its processivity is more sensitive to load, suggesting that the gating mechanisms that their control processivity may differ. To understand the motor roles that front-head gating and rear-head gating, we carried out stopped flow kinetics experiments using mant nucleotides, steady state assays, and single-molecule investigations to characterize the entire kinetic cycle a functional mouse KIF3A homodimer that exhibits similar motility to full-length KIF3A/B. Upon first encounter with the microtubule lattice, the motor exchanges mADP with an on-rate of 18 μM-1 s-1 and an off-rate of 27 s-1. When AMPPNP was used to entrap the motor in a two-head bound state, exchange kinetics were unchanged, indicating that rearward strain in the two-head-bound state does not alter nucleotide binding to the front head. Similar lack of front-head gating was found with mATP and when the neck linker domain was shortened from 17 to 14 residues to enhance intramolecular strain. In contrast, microtubule pelleting and single-molecule microscopy assays found that in ADP the motor dissociates with an off-rate of 2.1 s-1 and a KD of 0.5 μM, similar to its behavior in ATP. Hence, kinesin-2 processivity results from rear-head gating and not front-head gating. Based on the kinetics measurements, finally we propose a complete model of the kinesin-2 hydrolysis cycle that accounts for all of the kinetics and motility data. This study provides the direct evidence that rear-head gating does and not front-head gating does not play a role in kinesin-2 processivity and suggests that kinesin-2 mechanochemistry is specifically adapted for bidirectional transport.

The kinesin-5 KSP/Eg5 pauses at microtubule plus-ends and enhances tubulin polymerization by increasing the growth rate and decreasing the catastrophe rate (Chen et al. 2015). The end pausing resembles that of the kinesin-8, kip3p, but has the opposite effect on microtubule dynamics. Upon initial binding to the microtubule lattice, kinesin-5 was reported to pause for 1-second pause before stepping (Krzysiak et al. 2008). However, this “initial pause” predicted from biochemical experiments have not been observed in our single-molecule motility experiments. Compared to transport kinesins, kinesin-5 is less processive but paradoxically less load-dependent, suggesting that kinesin-5 may use different gating mechanisms during processive walking. We measured a kcat of 12 s-1 for the kinesin-5 monomer ATPase and a K0.5 of 0.9 μM, which generates a kbiATPase of 13 μM-1 s-1. The bimolecular microtubule-encounter rate kbiADP was 9 μM-1 s-1 which predicts a monomer chemical processivity of kbi ratio ∼1.4. This suggests that for each ATP turnover, kinesin-5 monomer readily detaches from the microtubule. The ADP off-rate of the bound head was 40 s-1, consistent with rate limiting hydrolysis. In half-site ADP release experiments with processive dimers, we measured maximal rates of 13 s-1 in ATP and 2.1 s-1 in ATPγS, which matched the single-molecule stepping rate in each nucleotide, suggesting that hydrolysis precedes attachment of the tethered head, consistent with recent work on kinesin-1. Hence, we propose that kinesin-5 processivity is gated by the race between detachment of the bound-head in the ADP-Pi state and attachment of the tethered-head. This study provides insights into gating mechanism with respect to processivity on the microtubule wall and in the possible strong-binding states at the microtubule plus-end.