Figure 3.

Separate CSPP1 domains control the balance between MT polymerization and depolymerization. (A) Schematic representation of the different CSPP1 constructs used and a summary of their MT binding ability and their effects on MT dynamics. Black boxes represent α-helical domains larger than 20 amino acids predicted by AlphaFold; asterisks indicate previously unidentified helices. ++: frequently observed at protein below 40 nM; +/−: occasionally observed at protein concentrations below 40 nM and/or frequently observed at protein concentrations up to 100 nM; - -: observed infrequently or not observed at all even at protein concentrations higher than 100 nM. MTB, MT binding domain; PD, pausing domain. (B) Histograms of fluorescence intensities of single GFP molecules immobilized on coverslips and GFP-CSPP-L molecules upon initial binding to the MT lattice in a separate chamber on the same coverslip (symbols) and the corresponding fits with lognormal distributions (lines). Number of molecules in analysis: single GFP, n = 16,795; GFP-CSPP-L, n = 54 (for the latter, initial binding events were manually selected for analysis) from one representative experiment. Dashed lines show corresponding modal values. (C) Quantification of the fluorescence intensity of GFP-CSPP-L molecules over time after initial binding. Green line shows one-phase association fit. Grey horizontal lines correspond to the quartile values of fluorescence intensities of single GFP in the histogram shown in B. Number of molecules in analysis; GFP-CSPP-L, n = 54 from one representative experiment. Error bars represent SEM. (D) Kymographs illustrating the behavior of the indicated GFP-CSPP fragments on the MT lattice in presence of 20 nM mCherry-EB3 and 15 µM rhodamine-tubulin. Scale bars, 2 μm (horizontal) and 30 s (vertical). (E) Kymographs illustrating MT growth with 20 nM mCherry-EB3 alone or together with 10 or 100 nM of the indicated GFP-CSPP1 constructs. Scale bars, 2 μm (horizontal) and 60 s (vertical). (F and G) Parameters of MT plus end dynamics in the presence of 20 nM mCherry-EB3 together with 10 or 100 nM of the indicated GFP-CSPP1 constructs (from kymographs as shown in B). Events were classified as pauses when the pause duration was longer than 20 s. Total number of growth events, pauses, and MTs analyzed (C); EB3 alone, n = 514, 0, 53; EB3 with 10 nM CSPP-L, n = 731, 518, 89; EB3 with 10 nM H4+LZ, n = 855, 0, 87; EB3 with 100 nM H4+LZ, n = 987, 0, 103; EB3 with 10 nM MTB+LZ, n = 1,006, 0, 109; EB3 with 100 nM MTB+LZ, n = 1,206, 0, 139; EB3 with 10 nM MTB+LZ+PD, n = 934, 0, 104; EB3 with 100 nM MTB+LZ+PD, n = 776, 707, 123. Total number of transition events analyzed (D): EB3 alone, n = 461, 0, 0, 0, 4, 0; EB3 with 10 nM CSPP-L, n = 24, 465, 455, 22, 25, 21; EB3 with 10 nM H4+LZ, n = 751, 0, 0, 0, 3, 0; EB3 with 100 nM H4+LZ, n = 889, 0, 0, 0, 15, 0; EB3 with 10 nM MTB+LZ, n = 902, 0, 0, 0, 26, 0; EB3 with 100 nM MTB+LZ, n = 1,035, 0, 0, 0, 582; EB3 with 10 nM MTB+LZ+PD, n = 797, 0, 0, 0, 191, 0; EB3 with 100 nM MTB+LZ+PD, n = 126, 545, 520, 105, 107, 121. Bars for growth rate and pause duration represent pooled data from three independent experiments. For dynamic state and transition frequencies, bars represent the average of the means (symbols) of three independent experiments. Error bars represent SEM. ***, P < 0.001; n.s., not significant; Kruskal–Wallis test followed by Dunn’s post-test. Data for EB3 alone and EB3 with 10 nM CSPP-L is the same as in Fig. 1, E and F. See also Fig. S2.