Vermeulen BJ et al. (2024), γ-TuRC asymmetry induces local protofilament mi...

XB-ART-60663

EMBO J 2024 Apr 10; doi: 10.1038/s44318-024-00087-4.

Show Gene links Show Anatomy links

γ-TuRC asymmetry induces local protofilament mismatch at the RanGTP-stimulated microtubule minus end.

Vermeulen BJ , Böhler A , Gao Q , Neuner A , Župa E , Chu Z , Würtz M , Jäkle U , Gruss OJ , Pfeffer S , Schiebel E .

???displayArticle.abstract???
The γ-tubulin ring complex (γ-TuRC) is a structural template for de novo microtubule assembly from α/β-tubulin units. The isolated vertebrate γ-TuRC assumes an asymmetric, open structure deviating from microtubule geometry, suggesting that γ-TuRC closure may underlie regulation of microtubule nucleation. Here, we isolate native γ-TuRC-capped microtubules from Xenopus laevis egg extract nucleated through the RanGTP-induced pathway for spindle assembly and determine their cryo-EM structure. Intriguingly, the microtubule minus end-bound γ-TuRC is only partially closed and consequently, the emanating microtubule is locally misaligned with the γ-TuRC and asymmetric. In the partially closed conformation of the γ-TuRC, the actin-containing lumenal bridge is locally destabilised, suggesting lumenal bridge modulation in microtubule nucleation. The microtubule-binding protein CAMSAP2 specifically binds the minus end of γ-TuRC-capped microtubules, indicating that the asymmetric minus end structure may underlie recruitment of microtubule-modulating factors for γ-TuRC release. Collectively, we reveal a surprisingly asymmetric microtubule minus end protofilament organisation diverging from the regular microtubule structure, with direct implications for the kinetics and regulation of nucleation and subsequent modulation of microtubules during spindle assembly.

???displayArticle.pubmedLink??? 38600243
???displayArticle.link??? EMBO J
???displayArticle.grants??? [+]

Species referenced: Xenopus laevis
Genes referenced: dtx1 emd grap2 grip2
GO keywords: microtubule [+]
???displayArticle.antibodies??? Tubg1 Ab6

???attribute.lit??? ???displayArticles.show???

	Synopsis The universal microtubule nucleator γ-TuRC assumes a microtubule-incompatible open conformation prior to nucleation. This work finds it transitioning to a partially closed conformation, with local misalignments to the nucleated microtubule in the native RanGTP-dependent nucleation pathway. • Cryo-EM analysis describes the structure of the γ-TuRC-capped microtubule minus end natively nucleated through the RanGTP pathway in Xenopus laevis egg extracts. • The γ-TuRC conformation approaches but does not reach microtubule lattice symmetry after microtubule nucleation. • The associated microtubule minus end is asymmetric and some of its protofilaments are misaligned with the partially closed γ-TuRC. • The architecture of the γ-TuRC and the attached microtubule facilitates selective binding of microtubule-modulating protein CAMSAP2 at the γ-TuRC-capped microtubule minus end.
	Figure 1. Cryo-EM characterisation of γ-TuRC-capped MT minus ends purified from X. laevis egg extract. (A) Examples of γ-TuRC-capped MT minus ends identified on cryo-EM micrographs. Scale bar represents 20 nm. (B) Representative 2D classes of capped MT minus ends (mask diameter 38 nm). (C) Example raw micrograph cut-out (labelled as raw) and corresponding Moiré pattern (labelled as filtered) for a 13 protofilament (pf) and 14 protofilament MT (obtained from a non-capped MT). Arrows indicate Moiré pattern transitions characteristic of 14 protofilament MTs. (D) Fraction of Paclitaxel-stabilised, γ-TuRC-capped MTs with a specific protofilament number, as determined by Moiré pattern analysis and computational particle sorting on a subset of capped MT minus ends from n = 33 individual MTs. Error bars indicate standard deviation centred around the fraction of MTs. Source data are available online for this figure.
	Figure EV1. Cryo-EM processing workflow. (A) Detailed image processing scheme. Colours of text and reconstructions indicate the particles used for reconstruction; capped MT minus ends in green, isolated γ-TuRCs in red, a mixture of both in khaki and reference densities in yellow. Respective Fourier Shell Correlation (FSC) curves for resolution values marked by orange and green line are shown in panel (D). (B) Sample cryo-EM micrograph of γ-TuRC-capped MTs from Xenopus laevis, with particles retained for final reconstruction indicated. Scale bar represents 50 nm. (C) Comparison of the reconstruction of the γ-TuRC at the MT minus end (orange, left) and the reconstruction of isolated γ-TuRC particles (grey, right) from the same joint refinement run, showing the former reconstruction is not biased by supplementation of isolated γ-TuRC during refinement. Outlined box shows the presence of density for the GRIP2 domain and γ-tubulin of spokes 5 and 6 (indicated by a box) in the γ-TuRC at the MT minus end. This density was subtracted from the supplemented particle images of isolated γ-TuRC and removed from the isolated γ-TuRC reconstruction that served as a reference for refinement. To enable a fair comparison, both reconstructions are low-pass filtered to 17 Å, the resolution of the reconstruction for the γ-TuRC at the MT minus end. Spoke numbering and colouring scheme are indicated. Density for spokes in the background was hidden (clipped) for visual clarity in both panels. (D) FSC curve for the γ-TuRC at the MT minus end from γ-TuRC-focused refinement, after removal of supplemented particles (orange) and FSC curve for the MT-dominated local refinement after γ-TuRC-focused refinement (green). Both curves were calculated taking into account a corrected, unbinned pixel size of 2.54 Å/px (see Methods). Threshold at FSC = 0.143 is indicated by a dashed line; the resolution at which the FSC curve first passes the threshold is specified. Source data are available online for this figure.
	Figure EV2. Refinement and reconstruction of the MT-capping γ-TuRC without particle supplementation confirms a partially closed conformation. (A–C) Refinement of only MT-capping-γ-TuRC particles at global angular sampling, without any particle supplementation, following refinement with particle supplementation. Atomic models for the hypothetical closed γ-TuRC (Kollman et al, 2015) (A), the partially closed γ-TuRC (this study, B) and the open γ-TuRC (PDB 6TF9 (Liu et al, 2020), C) were then rigid-body docked based on the GRIP1/2 domains and γ-tubulin molecules of spokes 3 to 9. The partially closed γ-TuRC (B) fits the reconstruction well, in spite of the absence of supplemented particles during the refinement, as opposed to the closed (A) and open (C) γ-TuRC, which clearly deviate from the reconstruction, especially at spokes 11 to 14 (highlighted by a grey line). This shows that the particle supplementation does not induce bias towards the open conformation. For clarity, only a slice through the reconstruction is shown, superposed with γ-tubulin molecules 2–14 of the fit model. The slice was chosen to highlight the fit in the upper half of the spokes. As the slice does not cover spokes 2 and 3, they are shown transparent. Spoke numbering is indicated.
	Figure EV3. The conformation of the γ-TuRC and attached MT is not affected by the identity of the MT stabilisation agent. (A) Reconstruction obtained from γ-TuRC-focused refinement of MT-capping γ-TuRCs with only DTX-stabilised particles (top) or only Paclitaxel-stabilised particles (bottom). A model for the γ-TuRC was generated for each reconstruction separately by spoke-wise rigid body docking. A high confidence threshold is used to highlight the fit at the level of γ-tubulin. Spoke numbering is indicated. Colouring as in Fig. 2A. (B–E) Euclidian translation distance (B), downward change in helical pitch (C), translation towards the MT axis (D) and rotation around the MT axis (E) required to convert the models to the hypothetical, fully closed γ-TuRC for each spoke. Parameters measured from the centre of mass of γ-tubulin. (F) Reconstructions of MT-dominated refinements of γ-TuRC-capped MTs with only DTX-stabilised (green) or only Paclitaxel-stabilised particles (white). Density for spokes in the background was hidden for visual clarity. Colouring scheme is indicated. (G) Coordinates of the centre of mass for the first α-tubulin in each protofilament in the plane orthogonal to the MT axis for regular 13 protofilament lattice (grey), Paclitaxel-stabilised γ-TuRC-capped MT (dark blue) and DTX-stabilised γ-TuRC-capped MT (light blue). Protofilaments are numbered by the respective spokes in the γ-TuRC. The MT axis is placed on the origin, (0,0). Colouring scheme is indicated. Source data are available online for this figure.
	Figure 2. The γ-TuRC assumes a partially closed conformation at the minus end of MTs nucleated through the native RanGTP pathway. (A) Cryo-EM reconstruction of the γ-TuRC at the MT minus end. Density for the MT is shown in white. (B) Spoke-wise rigid-body fit of the atomic model for the Xenopus laevis γ-TuRC (PDB 6TF9) (Liu et al, 2020). Colouring scheme and spoke numbers are indicated. (C) Comparison of the γ-TuRC at the MT minus end with the isolated γ-TuRC (PDB 6TF9) (Liu et al, 2020) and the hypothetical fully closed γ-TuRC, generated based on (EMD 2799) (Kollman et al, 2015). View along the longitudinal axis. Outlines of molecular surfaces are shown for γ-tubulins of spokes 2–14; conformational change from the open to the hypothetical closed γ-TuRC at spoke 14 is indicated by an arrow. (D) Euclidean distance separating γ-tubulin positions in the isolated γ-TuRC (dark grey) or the MT-capping γ-TuRC (orange) from their positions in the hypothetical, fully closed γ-TuRC conformation for each spoke. Distance measured between the centres of mass of γ-tubulins. Colouring as in (C). (E) Zoom-in of the lumenal bridge in the reconstruction of the MT-capping γ-TuRC, superposed with the atomic model of the isolated X. laevis γ-TuRC (PDB 6TF9) (Liu et al, 2020). While the MZT1/GCP6N and MZT1/GCP3N modules of the lumenal bridge are well resolved, density for actin is reduced. Colouring as indicated; components outside the lumenal bridge are shown in grey. Source data are available online for this figure.
	Figure 3. Capped MT minus ends purified from X. laevis egg extract are asymmetric and feature breaks in inter-protofilament and protofilament-γ-tubulin contacts. (A) Cryo-EM reconstruction of the γ-TuRC-capped MT minus end, resulting from MT-dominated refinement. (B) Atomic models of a protofilament-wise fitted MT and the partially closed γ-TuRC (Fig. 2B) displayed in the density. Numbering is shown for γ-TuRC spokes; colour scheme is indicated at the bottom. (C) Distance from the MT axis for each protofilament for the MT at capped minus ends (green) or an ideal 13 protofilament MT (grey). (D) Molecular surfaces are shown for γ-tubulins (filled) and the first α-tubulin (outlined) of each protofilament. View along the MT axis; γ-tubulin of spoke 1 was omitted for clarity. (E) Lateral offsets (perpendicular to the MT axis) between γ-tubulin and the first α-tubulin of the respective protofilament. Parameters in (C) and (E) are measured from the respective subunit’s centre of mass. (F) Left: Surface representation of the MT seam in a typical 13 protofilament MT, where lateral contacts form between α- and β-tubulin (PDB 6EW0 (Manka and Moores, 2018)). Right: Surface representation of the seam location in the minus end of capped MTs. Dashed boxes indicate the location of the seam (left) or where the seam will form (right) further along the length of the MT or after potential release from the γ-TuRC. Colouring scheme as in panel (B). Source data are available online for this figure.
	Figure EV4. Not all MT protofilaments are equally rigid with respect to the γ-TuRC. (A, B) Refinement focused on the γ-TuRC, in which all γ-TuRC spokes are well-defined, indicating that the γ-TuRC was well-aligned all around. In contrast, the MT density shows local variations in definition: protofilaments associated with spokes 7–11 of the γ-TuRC (A) are considerably better defined than those associated with spokes 2–5 (B). This indicates increased conformational plasticity of the MT relative to the γ-TuRC at the symmetric side of the γ-TuRC. Reconstructions are local resolution-filtered, shown at low threshold to emphasise MT density. Spoke numbers are indicated. Inset schematics show the orientation of the γ-TuRC.
	Figure EV5. The conformation of the γ-TuRC and attached MT is not affected by the shortening procedure applied during purification. (A) Reconstruction obtained from γ-TuRC-focused refinement of γ-TuRCs capping Paclitaxel-stabilised microtubules which were (top) or were not (bottom) subjected to the shortening procedure during purification. A model for the γ-TuRC was generated for each reconstruction separately by spoke-wise rigid body docking. A high confidence threshold is used to highlight the fit at the level of γ-tubulin. Spoke numbering is indicated. Colouring as in Fig. 2A. (B–E) Euclidian translation distance (B), downward change in helical pitch (C), translation towards the MT axis (D) and rotation around the MT axis (E) required to convert the models to the hypothetical, fully closed γ-TuRC for each spoke. Parameters measured from the centre of mass of γ-tubulin. (F) Reconstructions obtained from MT-dominated refinements of γ-TuRC-capped Paclitaxel-stabilised MTs which were (white) or were not (purple) subjected to the shortening procedure during purification. Density for spokes in the background was hidden for visual clarity. Colouring scheme is indicated. (G) Coordinates of the centre of mass for the first α-tubulin in each protofilament in the plane orthogonal to the MT axis for the regular 13 protofilament lattice (grey), the Paclitaxel-stabilised γ-TuRC-capped MT where shortening was (dark blue) or was not (green) applied during purification. Individual protofilaments could not be fitted at spokes 4 to 6 for the condition without shortening due to lower density quality. Protofilaments are numbered by the respective spokes in the γ-TuRC. The MT axis is placed on the origin, (0,0). Colouring scheme is indicated. Source data are available online for this figure.
	Figure 4. CAMSAP2 specifically recognises the minus end of γ-TuRC-capped MTs nucleated through the RanGTP pathway. (A) Exemplary MTs nucleated through the RanGTP pathway in X. laevis egg extract that show colocalisation of the γ-TuRC (yellow) and GFP-CAMSAP2 (red) at the minus end of MTs (blue), as judged by fluorescence microscopy. Scale bar represents 1 μm. (B) Normalised average intensity of CAMSAP2 (red), γ-TuRC (yellow) and MT signals (blue) along the length of γ-TuRC-capped MTs (n = 69). Only MTs longer than 5 μm were considered; data were combined from three biological replicates. (C) Average intensity of CAMSAP2 signal within 0.6 μm of the γ-TuRC signal peak (n = 552 from 69 MTs, of which 28, 19 and 22 for each respective replicate) or further away (n = 823 from 69 MTs). Intensities were normalised to the mean CAMSAP2 intensity on the MT lattice for each individual replicate. p = 3.2 × 10−68 Dots indicate mean values of the three biological replicates. ***p < 0.001 by a one-tailed Welch’s t-test. Data in (B) and (C) shown as mean with 95% confidence interval for all individual data points. Source data are available online for this figure.
	Figure 5. Possible roles for partial γ-TuRC closure and MT misalignment in the regulation of MT nucleation and release. (A) The open γ-TuRC (dark grey curve) displays highly cooperative nucleation behaviour in response to α/β-tubulin concentration but has a low base rate of nucleation (Thawani et al, 2020). The fully closed γ-TuRC (light grey) has a high base rate of nucleation with a lack of cooperativity. A partially closed γ-TuRC (orange) would likely have an intermediate base rate of nucleation, while still retaining some cooperativity. Such behaviour could aid synergy of the γ-TuRC with α/β-tubulin-enriching factors. Graphs are for illustrative purposes and do not represent experimental or simulated data. (B) Misalignment between the γ-TuRC and MT protofilaments may promote release of freshly nucleated MTs from the γ-TuRC, e.g. by CAMSAP2 (red), and recycling of nucleation-competent γ-TuRC. Spoke 1 and 14 are coloured to highlight different degrees of γ-TuRC closure. Spoke numbers are indicated for spokes 1 and 14. Missing interactions around the protofilament at spoke 2 highlighted with red outlines. (C) Different activation cues may induce different degrees of γ-TuRC closure to accommodate differential requirements of nucleation rate and minus end stability. (D) The γ-TuRC may transiently visit a fully closed conformation that favours MT nucleation and subsequently relax to the partially closed conformation observed in our study to favour MT release.
	Appendix Figure S1 - Purification of γ-TuRC-capped MT minus ends from Xenopus laevis egg extract. A Schematic representation of the workflow for purification of γ-TuRC-capped MTs from Xenopus laevis egg extract. B Sample fluorescence microscopy images used for judging aster-forming capacity of Xenopus laevis egg extracts, a prerequisite for further purification of γ-TuRC-capped MT minus ends. Scale bars represent 100 µm in full images and 25 µm in zoomed cut-outs.
	Appendix Figure S2 - RanGTP treatment drastically increases the number of γ-TuRC-capped MTs isolated from X. laevis egg extract. A,B Representative negative stain EM micrograph at the grid square level (top left, scale bars 3 µm), at intermediate magnification (top right, scale bar 200 nm) and a gallery of cut-outs at high magnification (bottom, scale bar 20 nm) of γ-TuRC-capped MTs purified with (A) or without (B) RanGTP induction. MTs were stabilised with Paclitaxel.
	Appendix Figure S3 - Cryo-EM reconstruction of the MT lattice distant from the capped minus end. A Overview of the MT lattice density of Paclitaxel-stabilised MTs distant from the capped minus end, clearly resolving defined secondary structure elements, indicating the MT lattice is structurally intact. Dashed box indicates the area focused on in panel (B). B Dimer-wise fit of an atomic model of mammalian α- and β-tubulin (PDB 6EW0 (Manka & Moores, 2018)) into the density shown in (A), viewed from the MT surface (left) and lumen (right). C Fourier Shell Correlation (FSC) curve used for determining the resolution (as specified in the plot) of the reconstruction in panel (A) at threshold FSC=0.143 (indicated by a dashed line). Source data is available for this figure.
	Appendix Figure S4 - Detailed structural analysis of the γ-TuRC at the MT minus end. A-C Rotation around the MT axis (A), translation towards the MT axis (B) and downward change in helical pitch (C) required to convert the isolated γ-TuRC (dark grey) or the γ-TuRC at the MT minus end (orange) to the hypothetical, fully closed γ-TuRC for each spoke. Inset schematics represent the quantified parameters. Parameters measured from the center of mass of γ-tubulin. D,E Comparison of the γ-TuRC at the MT minus end with the isolated γ-TuRC (PDB 6TF9(Liu et al, 2020), (D)) and the hypothetical fully closed γ-TuRC (E), generated from EMD 2799 (Kollman et al, 2015) (see Methods). Lumenal bridge components are omitted. Molecular surfaces generated from atomic models using the molmap function in UCSF ChimeraX (Goddard et al, 2018). F Zoom-in of the lumenal bridge in the reconstruction of the isolated γ-TuRC (left panel) and the γTuRC at the MT minus end (right), both superposed with the atomic model of the isolated X. laevis γTuRC (PDB 6TF9) (Liu et al., 2020). Difference density (red, middle panel) superposed with the reconstruction of the γ-TuRC at the MT minus end (grey) highlights that the MZT1/GCP6N and MZT1/GCP3N modules are well resolved, while density for actin is significantly reduced. All reconstructions were low-pass filtered to 17 Å. Colouring as indicated; components outside the lumenal bridge are shown in grey. Source data is available for this figure.
	Appendix Figure S5 - Extended analysis of the minus end of capped MTs. A Quantification of the lateral distance between the first α-tubulin in the protofilament at spoke N to the first α-tubulin in the protofilament at spoke N+1. B Quantification of the helical pitch increment for the protofilament at spoke N compared to the protofilament at spoke 2. Colouring as in panel (A). All parameters were calculated using the center of mass of the α-tubulin molecule closest to the γ-TuRC, with schematics illustrating the plotted parameter. C Zoom-in on the density connecting the protofilament at spoke 2 with spoke 1 and 14 (highlighted by dashed circle) in the MT-dominated reconstruction of the γ-TuRC-capped MT. Spoke numbering is indicated. D The model of the γ-TuRC-capped MT generated in this study (white and dark grey) as well as spokes 13 and 14 with the terminal MZT1/GCP3N module of the human recombinant γ-TuRC (PDB 7QJD (Wurtz et al, 2022)) superposed to the reconstruction shown in panel (C). Spokes 13, 14 and the MZT1/GCP3N module were docked as a rigid body. Colouring scheme indicated. Source data is available for this figure.
	Appendix Figure S6 - Cryo-EM processing workflow for γ-TuRC-capped MTs not subjected to the shortening procedure. Detailed image processing scheme. Colours of text and reconstructions indicate the particles used for reconstruction; capped MT minus ends in green, isolated γ-TuRCs in red, a mixture of both in khaki and reference densities in yellow. Densities are shown for schematic purposes.
	Appendix Figure S7 - Binding of CAMSAP2 to the lattice and γ-TuRC-capped minus ends of MTs nucleated through the RanGTP pathway in X. laevis egg extract. A Fraction of γ-TuRC-capped MTs with CAMSAP2 binding to the minus end observed by multi-colour fluorescence microscopy (within 1 μm of the MT end; n=69, three biological replicates). Example fluorescence microscopy images are shown in Figure 4A. B Fraction of γ-TuRC-capped MTs and uncapped MTs with CAMSAP2 binding to either end observed by multi-colour fluorescence microscopy (within 1 μm of the MT end; n=69 for capped MTs, n=60 for uncapped MTs, p=0.000418, three biological replicates). C Fraction of γ-TuRC-capped MTs and uncapped MTs with CAMSAP2 binding to the MT lattice observed by multi-colour fluorescence microscopy (i.e., >1 μm from either MT end; n=69 for capped MTs, n=60 for uncapped MTs, p=0.3181). Dots in (A), (B) and (C) indicate mean values of the three biological replicates; data are shown as mean with 95% confidence interval for all individual data points. *** p<0.001, n.s. non-significant. Significance determined using a one-tailed Welch’s t-test. D, E Immunogold labelling of γ-TuRC-capped MTs isolated from X. laevis egg extract incubated with GFP-CAMSAP2. GFP-CAMSAP2 and γ-tubulin both localise to the same end of γ-TuRC-capped MTs, which is not observed in samples where GFP-CAMSAP2 is absent (E). Red (CAMSAP2) and yellow (γ-TuRC) asterisks indicate gold beads at the MT minus end. Additional density at MT ends may be attributed to staining of antibodies as well as potential accumulation of condensates of CAMSAP2 (Imasaki et al, 2022). 9 F γ-TuRC-capped MT minus ends isolated from X. laevis egg extract were incubated with GFPCAMSAP2 and labelled with single gold conjugate: anti-GFP antibody labelled with 10 nm Protein A gold conjugated to rabbit anti-goat antibody (left), anti-γ-tubulin antibody labelled with rabbit antimouse antibody that was subsequently labelled with 15 nm Protein A gold (middle) and labelling of only rabbit anti-mouse antibody with 15 nm Protein A gold (right). Red (CAMSAP2) and yellow (γTuRC) asterisks indicate gold beads at the MT minus end. Dots in legend indicate approximate observed size of respective gold beads. Scale bar in (D), (E) and (F) represents 25 nm. Source data is available for this figure.
	Appendix Figure S8 - Binding of CAMSAP2 to the lattice and γ-TuRC-capped minus ends of MTs nucleated in vitro. A 2D classes obtained from negative stain EM imaging of the recombinant γ-TuRC preparation used for in vitro MT nucleation and CAMSAP2 binding experiments. Scale bar represents 25 nm. B Example negative stain EM micrograph of the recombinant γ-TuRC preparation used for in vitro MT nucleation and CAMSAP2 binding experiments. A small number of exemplary γ-TuRC particles are indicated. Scale bar represents 50 nm. C 3D reconstruction of well-aligning γ-TuRC particles showing 13 stoichiometric and slightly underrepresented spoke 14, likely due to its known conformational plasticity. Subsequent focused classification attempts on spoke 13 and 14 with multiple T-factors revealed no classes indicative of a 11 population of γ-TuRC with 12 or fewer spokes. Atomic model of the recombinant γ-TuRC (PDB 7QJC) is fit. Actin is displayed in red, MZT1 in pink, otherwise, colouring as in Fig. 2B. Spoke numbers are indicated. D Normalised average intensity of CAMSAP2 (red), γ-TuRC (yellow) and MT signals (blue) along the length of γ-TuRC-capped MTs observed by multi-colour fluorescence microscopy (n=65). Curves were normalised to peak at 1. Only MTs longer than 5 μm were considered. Data were combined from three biological replicates. E Gallery of examples illustrating CAMSAP2 (red) colocalising with the γ-TuRC (yellow) and binding to the lattice of in vitro nucleated MTs (blue). Scale bar represents 2 µm. F Average intensity of CAMSAP2 signal within 0.75 μm of the γ-TuRC signal peak (n=795 from 159 MTs, of which 48, 53 and 58 for each respective biological replicate) or further away (n=3068 from 159 MTs). Intensities were normalised to the mean CAMSAP2 intensity on the MT lattice for each biological replicate. p=4.0x10-49. G Fraction of γ-TuRC-capped MTs with CAMSAP2 binding to the minus end (within 1 μm of the MT end; n=159). H Fraction of γ-TuRC-capped MTs and uncapped MTs with CAMSAP2 binding to either end (within 1 μm of the MT end; n=159 in both conditions, p=8.63x10-7 ). I Fraction of γ-TuRC-capped MTs and uncapped MTs with CAMSAP2 binding to the MT lattice (i.e., >1 μm from either MT end; n=159 in both conditions, p=0.05806). Dots in (F), (G), (H) and (I) indicate mean values of the three biological replicates; data are shown as mean with 95% confidence interval for all individual data points. *** p<0.001, n.s. non-significant. Significance determined using a one-tailed Welch’s t-test. Source data is available for this figure.