In vitro depolymerization in various nucleotide states and MCAK–tubulin stoichiometries. (A) MT depolymerization reactions performed for 10 min at 23°C with 0.5 mM of the indicated nucleotide and 2.5 μM GMPCPP-tubulin MTs, stopped by centrifugation. Enzyme concentration: 80 nM wt MCAK in ATP (1:31 MCAK/tubulin) and 510 nM in all other reactions (1:5). Tubulin release was measured by a pelleting assay using the stoichiometries of active motor to total tubulin indicated. Values are normalized to control (no motor) tubulin release of 0.13 nM per second (SD 0.02 nM). In contrast to wt MCAK, E491A releases more tubulin in ADP and ADP-Al-F than in ATP. (B) Time course of depolymerization as in A, except 1.7 μM GMPCPP tubulin, 21 nM wt MCAK in ATP (1:81 motor/tubulin), and 510 nM in all other reactions (1:3 motor/tubulin). E491A (at 510 nM) releases 1,200 nM tubulin dimers in 30 min in ADP, which suggests that the motor is recycling. In contrast, in ATP, less than equimolar tubulin is released in 30 min by E491A. (C) Linear depolymerization rates of MT ends directly observed in TIRF microscope. MTs tethered to a coverslip observed in a reaction chamber containing 1.25 μM of GMPCPP-Cy5-tubulin MTs, 2 mM of the indicated nucleotide, and 400–1,600 nM EGFP-E491A-MCAK. The rates are normalized to the nonmotor control tubulin loss rates that were observed to be 0.08 nM per second (SD, 0.03). E491A is capable of disassembling MTs at a high rate in ATP but only when the proportion of motor exceeds that of tubulin. At ∼1:3 motor/tubulin, E491A depolymerizes MTs at a higher rate in ADP than in ATP. (D) Loss of MT polymer observed by light scattering at very low stoichiometry (1:117 motor/tubulin or 1.4:1 MCAK/protofilament ends). Turbidity of reaction mixes with 0.2 mM of the indicated nucleotide, 14 nM MCAK, and 1.64 μM GMPCPP tubulin as 2.5-μm mean length MTs (10.6 nM protofilament ends) at 23°C and monitored at 340 nm. The rate of tubulin loss in the absence of motor and free tubulin is 0.10 nM per second (SD, 0.02), which is consistent with the control rates of tubulin loss for the pelleting and TIRF assays. At <1:100 motor per tubulin stoichiometry, E491A is incapable of disassembling MTs in ATP, whereas in ADP, ∼18 moles of tubulin were released per mole of E491A, indicating that the motor is recycling during successive rounds of tubulin removal. pf, protofilament. Error bars indicate SD of the measured quantity (tubulin released or the linear depolymerization rate).