Figure S2. Effect of Lis1, temperature,... | Rockefeller University Press

Figure S2.

$Effect of Lis1, temperature, and adaptor identity on dynein motility. (A) Representative kymographs showing the motility of mEGFP-dynein on surface-immobilized Atto647N-labeled GMPCPP-microtubules in the presence of dynactin and BicD2N1–400 at different mCherry-Lis1 concentrations. Concentrations as indicated. Lis1 stimulates BicD2N1–400-induced processive motility in a dose-dependent manner. (B) Velocity distributions of mEGFP-dynein processive runs (mean ± SEM) in the presence of either mScarlet-NuMAN-term, mScarlet-NuMAFL or BicD2N1–400, as indicated. Each circle corresponds to the velocity of a single processive segment of a run; n = 49, 55, 50, 72, 166 velocities; adjusted P values by Welch’s ANOVA test with Holm-Sidak’s post-hoc test for multiple comparisons: 0.8246 (N-term versus FL, 30°C), 0.8429 (N-term versus BicD2N1–400, 30°C), 0.7750 (FL versus BicD2N1–400, 30°C); P value by Welch’s t test: 0.7228 (N-term versus FL, 18°C). Experiments performed with NuMA contained 3 nM mEGFP-dynein, 7 nM dynactin, 50 nM NuMA, 650 nM Lis1; experiments with BicD2N1–400 contained 7 nM mEGFP-dynein, 14 nM dynactin, 200 nM BicD2N1–400, 100 nM Lis1. Velocities measured at 18°C were approximately threefold lower than those measured at 30°C, consistent with the temperature effect previously reported (Hong et al., 2016; Ruhnow et al., 2017). (C and D) Comparison of run lengths for dynein activated by NuMAN-term versus BicD2N1–400 (C) and by NuMAN-term versus NuMAFL (D). Left: survival probability (1-CDF) of all measured run lengths for NuMAN-term (n = 204) and BicD2N1‒400-activated (n = 206) complexes. A large fraction of these processive events reach the microtubule end (NuMAN-termn end = 57; BicD2N1–400n end = 72); center: ratio of end-reaching events to total number of processive events per microtubule versus microtubule length; right: corrected survival probability of run length using the Kaplan-Meier estimator with end events being treated as right censored data points (Ruhnow et al., 2017). Survival probabilities are fitted with exponential function to estimate the run length; errors are approximated by ∆R=2R/N; dotted lines indicate the 95% CI of the survival probability. Experiments in C were performed at 7 nM mEGFP dynein, 14 nM dynactin, 100 nM mCherry-Lis1, and 200 nM AF546-NuMAN-term or BicD2N1‒400. Experiments in D were performed at 3 nM mEGFP dynein, 7 nM dynactin, 650 nM Lis1, and 50 nM AF546-NuMAN-term or mScarlet-NuMAFL. All experiments were carried out in dynein microscopy buffer. (E) Representative TIRF microscopy images showing the binding of 50 nM mScarlet-NuMAFL to surface-immobilized Atto647N-labeled GMPCPP-microtubules in BRB80 (containing 80 mM Pipes, as in NuMA microscopy buffer) and BRB20 (containing 20 mM Pipes, as in dynein microscopy buffer), supplemented by 60 mM KCl. At 80 mM Pipes, NuMA binds all along the GMPCPP microtubules, as shown in Fig. 2 C. At 20 mM Pipes, NuMA’s solubility is reduced, as shown by numerous NuMA aggregates in solution, which impacts its ability to bind microtubules.$

Effect of Lis1, temperature, and adaptor identity on dynein motility. (A) Representative kymographs showing the motility of mEGFP-dynein on surface-immobilized Atto647N-labeled GMPCPP-microtubules in the presence of dynactin and BicD2N^1–400 at different mCherry-Lis1 concentrations. Concentrations as indicated. Lis1 stimulates BicD2N^1–400-induced processive motility in a dose-dependent manner. (B) Velocity distributions of mEGFP-dynein processive runs (mean ± SEM) in the presence of either mScarlet-NuMA^N-term, mScarlet-NuMA^FL or BicD2N^1–400, as indicated. Each circle corresponds to the velocity of a single processive segment of a run; n = 49, 55, 50, 72, 166 velocities; adjusted P values by Welch’s ANOVA test with Holm-Sidak’s post-hoc test for multiple comparisons: 0.8246 (N-term versus FL, 30°C), 0.8429 (N-term versus BicD2N^1–400, 30°C), 0.7750 (FL versus BicD2N^1–400, 30°C); P value by Welch’s t test: 0.7228 (N-term versus FL, 18°C). Experiments performed with NuMA contained 3 nM mEGFP-dynein, 7 nM dynactin, 50 nM NuMA, 650 nM Lis1; experiments with BicD2N^1–400 contained 7 nM mEGFP-dynein, 14 nM dynactin, 200 nM BicD2N^1–400, 100 nM Lis1. Velocities measured at 18°C were approximately threefold lower than those measured at 30°C, consistent with the temperature effect previously reported (Hong et al., 2016; Ruhnow et al., 2017). (C and D) Comparison of run lengths for dynein activated by NuMA^N-term versus BicD2N^1–400 (C) and by NuMA^N-term versus NuMA^FL (D). Left: survival probability (1-CDF) of all measured run lengths for NuMA^N-term (n = 204) and BicD2N^1‒400-activated (n = 206) complexes. A large fraction of these processive events reach the microtubule end (NuMA^N-termn end = 57; BicD2N^1–400n end = 72); center: ratio of end-reaching events to total number of processive events per microtubule versus microtubule length; right: corrected survival probability of run length using the Kaplan-Meier estimator with end events being treated as right censored data points (Ruhnow et al., 2017). Survival probabilities are fitted with exponential function to estimate the run length; errors are approximated by $∆ R = 2 R / \sqrt{N}$ ⁠; dotted lines indicate the 95% CI of the survival probability. Experiments in C were performed at 7 nM mEGFP dynein, 14 nM dynactin, 100 nM mCherry-Lis1, and 200 nM AF546-NuMA^N-term or BicD2N^1‒400. Experiments in D were performed at 3 nM mEGFP dynein, 7 nM dynactin, 650 nM Lis1, and 50 nM AF546-NuMA^N-term or mScarlet-NuMA^FL. All experiments were carried out in dynein microscopy buffer. (E) Representative TIRF microscopy images showing the binding of 50 nM mScarlet-NuMA^FL to surface-immobilized Atto647N-labeled GMPCPP-microtubules in BRB80 (containing 80 mM Pipes, as in NuMA microscopy buffer) and BRB20 (containing 20 mM Pipes, as in dynein microscopy buffer), supplemented by 60 mM KCl. At 80 mM Pipes, NuMA binds all along the GMPCPP microtubules, as shown in Fig. 2 C. At 20 mM Pipes, NuMA’s solubility is reduced, as shown by numerous NuMA aggregates in solution, which impacts its ability to bind microtubules.