CAMSAP2 organizes a γ-tubulin-independent microtubule nucleation centre through phase separation - PubMed (original) (raw)

doi: 10.7554/eLife.77365.

Tsuyoshi Imasaki # 1 2 3, Shinsuke Niwa 4, Yumiko Saijo-Hamano 1, Hideki Shigematsu 5 6, Kazuhiro Aoyama 7 8, Kaoru Mitsuoka 8, Takahiro Shimizu 1, Mari Aoki 3, Ayako Sakamoto 3, Yuri Tomabechi 3, Naoki Sakai 5 6, Mikako Shirouzu 3, Shinya Taguchi 1, Yosuke Yamagishi 1, Tomiyoshi Setsu 1, Yoshiaki Sakihama 1, Eriko Nitta 1, Masatoshi Takeichi 9, Ryo Nitta 1 3

Affiliations

CAMSAP2 organizes a γ-tubulin-independent microtubule nucleation centre through phase separation

Tsuyoshi Imasaki et al. Elife. 2022.

Abstract

Microtubules are dynamic polymers consisting of αβ-tubulin heterodimers. The initial polymerization process, called microtubule nucleation, occurs spontaneously via αβ-tubulin. Since a large energy barrier prevents microtubule nucleation in cells, the γ-tubulin ring complex is recruited to the centrosome to overcome the nucleation barrier. However, a considerable number of microtubules can polymerize independently of the centrosome in various cell types. Here, we present evidence that the minus-end-binding calmodulin-regulated spectrin-associated protein 2 (CAMSAP2) serves as a strong nucleator for microtubule formation by significantly reducing the nucleation barrier. CAMSAP2 co-condensates with αβ-tubulin via a phase separation process, producing plenty of nucleation intermediates. Microtubules then radiate from the co-condensates, resulting in aster-like structure formation. CAMSAP2 localizes at the co-condensates and decorates the radiating microtubule lattices to some extent. Taken together, these in vitro findings suggest that CAMSAP2 supports microtubule nucleation and growth by organizing a nucleation centre as well as by stabilizing microtubule intermediates and growing microtubules.

Keywords: CAMSAP; E. coli; LLPS; TIRF; cell biology; cryo-EM; microtubule; molecular biophysics; mouse; nucleation; structural biology.

Plain language summary

Cells are able to hold their shape thanks to tube-like structures called microtubules that are made of hundreds of tubulin proteins. Microtubules are responsible for maintaining the uneven distribution of molecules throughout the cell, a phenomenon known as polarity that allows cells to differentiate into different types with various roles. A protein complex called the γ-tubulin ring complex (γ-TuRC) is necessary for microtubules to form. This protein helps bind the tubulin proteins together and stabilises microtubules. However, recent research has found that in highly polarized cells such as neurons, which have highly specialised regions, microtubules can form without γ-TuRC. Searching for the proteins that could be filling in for γ-TuRC in these cells some evidence has suggested that a group known as CAMSAPs may be involved, but it is not known how. To characterize the role of CAMSAPs, Imasaki, Kikkawa et al. studied how one of these proteins, CAMSAP2, interacts with tubulins. To do this, they reconstituted both CAMSAP2 and tubulins using recombinant biotechnology and mixed them in solution. These experiments showed that CAMSAP2 can help form microtubules by bringing together their constituent proteins so that they can bind to each other more easily. Once microtubules start to form, CAMSAP2 continues to bind to them, stabilizing them and enabling them to grow to full size. These results shed light on how polarity is established in cells such as neurons, muscle cells, and epithelial cells. Additionally, the ability to observe intermediate structures during microtubule formation can provide insights into the processes that these structures are involved in.

© 2022, Imasaki, Kikkawa et al.

PubMed Disclaimer

Conflict of interest statement

TI, SK, SN, YS, HS, KM, TS, MA, AS, YT, NS, MS, ST, YY, TS, YS, EN, MT, RN No competing interests declared, KA Kazuhiro Aoyama is affiliated with Thermo Fisher Scientific. The author has no financial interests to declare

Figures

Figure 1.

Figure 1.. Functional study of recombinant calmodulin-regulated spectrin-associated protein 2 (CAMSAP2).

(A) Schematic representation of the full-length CAMSAP2 constructs used in this study. CH, calponin-homology domain; MBD, microtubule-binding domain; CC, coiled-coil domain; CKK, C-terminal domain common to CAMSAP1 and two other mammalian proteins, KIAA1078 and KIAA1543. (B) (C) Size exclusion chromatography and SDS-PAGE of the peak fraction of (B) full-length CAMSAP2 and (C) GFP-CAMSAP2. (D) Total internal reflection fluorescence images of polarity-marked microtubules (magenta) decorated with purified GFP-CAMSAP2 (green). The minus-end segment of the microtubule is brighter than the plus-end segment.

Figure 1—figure supplement 1.

Figure 1—figure supplement 1.. Size exclusion chromatography with multi-angle light scattering of the calmodulin-regulated spectrin-associated protein 2-FL.

Figure 2.

Figure 2.. Calmodulin-regulated spectrin-associated protein 2 (CAMSAP2) stimulates microtubule nucleation.

(A) SDS-PAGE gels from a spin-down spontaneous nucleation assay of tubulin showing the total tubulin after 60 min of polymerization at 37°C (left gel). Polymerized tubulin was pelleted by centrifugation and then depolymerized on ice and centrifuged to remove debris, and the supernatant was subjected to SDS-PAGE (right gel). (B) SDS-PAGE gels from a spin-down spontaneous nucleation assay of tubulin with 1 µM CAMSAP2-FL showing the total tubulin-CAMSAP2 (left gel) and the polymerized/depolymerized tubulin (right gel). (C) Plots of the depolymerized tubulin concentrations determined by pelleting assay against the total tubulin concentrations determined by the reaction on SDS-PAGE gel from three independent assays (mean ± SD). The depolymerized tubulin from the tubulin assay is turquoise green, and that combined with CAMSAP2-FL is orange. The concentrations of tubulin greater than 0.1 µM are fitted with a trend line that has an x-intercept of 23.8 ± 0.96 µM (CcMT nucleation). (D) SDS-PAGE gels from a spin-down spontaneous nucleation assay of 10 µM tubulin with 10–1000 nM CAMSAP2-FL in the PEM (100 mM PIPES pH 6.8, 1 mM MgCl2, 1 mM EGTA, and 1 mM GTP) or PEM + 100 mM KCl. (E) Plots of the depolymerized pelleting assay in the different concentration of the CAMSAP2-FL with different buffer with SD from three independent assays (mean ± SD).

Figure 2—figure supplement 1.

Figure 2—figure supplement 1.. Comparison of pelleting assay.

Comparison of conventional pelleting assay without depolymerization (no deploy) and modified pelleting assay with depolymerization (depoly, which is the same data as shown in Figure 2C).

Figure 3.

Figure 3.. Calmodulin-regulated spectrin-associated protein 2 (CAMSAP2) forms co-condensate with tubulin in vitro.

(A) Intrinsic disorder prediction of CAMSAP2 by PONDR. CH, calponin-homology domain; MBD, microtubule-binding domain; CC, coiled-coil domain; CKK, C-terminal domain common to CAMSAP1 and two other mammalian proteins, KIAA1078 and KIAA1543. (B) Fluorescent image of GFP-CAMSAP2-FL condensates (top) and DIC image of CAMSAP2-FL condensates (bottom). (C, D) Phase diagram of 1 µM GFP-CAMSAP2-FL with indicated salt concentrations (C) and different concentrations of GFP-CAMSAP2-FL with 100 mM KCl (D). (E) Fusion of the GFP-CAMSAP2-FL condensate (also see Video 1). (F) Fluorescence recovery after photobleaching of GFP-CAMSAP2-FL condensates, acquired via confocal microscopy and (G) quantification. Time 00:00 (minutes:seconds) corresponds immediately after photobleaching. The graph shows the fluorescence recovery process of one of the four quantified droplets in Figure 3—figure supplement 1. (H) (I) GFP-CAMSAP2-FL and tubulin formed co-condensate in the physiological buffer (PEM with 100 mM KCl) and microtubule polymerization buffer (PEM).

Figure 3—figure supplement 1.

Figure 3—figure supplement 1.. Three different measurements of fluorescence recovery after photobleaching of GFP-calmodulin-regulated spectrin-associated protein 2-FL condensates, acquired via confocal microscopy and quantification.

Figure 4.

Figure 4.. Tubulin is incorporated into calmodulin-regulated spectrin-associated protein 2 (CAMSAP2) condensates to form aster-like structure.

(A) The procedure used to obtain the data in panels (B) and (C). GFP-CAMSAP2-FL, 10 μM tubulin, and 0.5 μM tetramethylrhodamine (TMR)-tubulin were mixed in BRB80 supplemented with 100 mM KCl and incubated for 10 min at 37°C. The solution was directly transferred onto a coverslip and observed by fluorescence microscopy. (B) Representative image of asters. GFP-CAMSAP2-FL (0.5 μM), tubulin (10 μM), and TMR-tubulin (0.5 μM) were co-incubated. (C) Quantification of the numbers of asters in solutions containing 10 μM tubulin, 0.5 μM TMR-tubulin, and 10, 25, 50, 100, 250, 500, and 1000 nM GFP-CAMSAP2-FL. The results of three independent assays are shown with dots. (D) Schematic showing reconstitution of CAMSAP2-containing foci. CAMSAP2 condensates were formed as described in Figure 3 and fixed on the coverslip by an anti-GFP antibody. (E) Soluble tubulins were incorporated into CAMSAP2 condensates within 1 min. (F) Schematic showing CAMSAP2-containing foci with soluble tubulin and GFP-CAMSAP2-FL. (G) CAMSAP2 in solution induced aster formation from CAMSAP2 condensates in a dose-dependent manner. Representative images for 0 and 50 nM GFP-CAMSAP2-FL are shown. (H) Quantification of microtubule formation from CAMSAP2 condensates. The percentages of CAMSAP2 condensates with microtubules among total CAMSAP2 condensates are shown. Each dot shows the results of three independent experiments. (I) Time-lapse images of aster formation. Tubulin (10 μM), TMR-tubulin (0.5 μM), and GFP-CAMSAP2-FL (50 nM) were incubated with CAMSAP2 condensates fixed on coverslips. Dynamic microtubules from CAMSAP2 condensates were observed (arrows). The scale bars indicate 5 µm. See Videos 3 and 4.

Figure 4—figure supplement 1.

Figure 4—figure supplement 1.. Calmodulin-regulated spectrin-associated protein 2 (CAMSAP2) localization in growing microtubule networks in HeLa cells.

Cells were monitored before nocodazole treatment (-nocodazole), just after nocodazole treatment (0 min), and 3 min and 35 min incubation at 37°C after washout. Immunostaining of intrinsic CAMSAP2 (magenta) and α-tubulin (green) in HeLa cells shows the local condensates of CAMSAP2 appeared before the initiation of microtubule polymerization. The scale bars indicate 5 µm.

Figure 5.

Figure 5.. Nucleation and aster formation activity of calmodulin-regulated spectrin-associated protein 2 (CAMSAP2).

Representative electron microscopy (EM) images are shown from at least three independent assays. (A) Negative stain EM micrographs of 1 µM CAMSAP2-FL incubated at 25°C for 30 min. (B) Negative stain EM micrographs of 10 µM tubulin polymerized with 1 µM of CAMSAP2-FL after incubation at 37°C for 10 min. Aster-like microtubule structures were observed. Negative stain micrographs of tubulin with CAMSAP2-FL incubated at various conditions were also available in Figure 5—figure supplement 2. (C) Cryo-EM micrographs of 30 µM tubulin polymerized with 3 µM CAMSAP2-FL after incubation at 37°C for 10 min captured at different magnifications. Cam2-asters are indicated by the red arrows. The cryo-EM micrographs of 30 µM tubulin on ice, polymerized for 1, 3, and 10 min at 37°C are available in Figure 5—figure supplement 3.

Figure 5—figure supplement 1.

Figure 5—figure supplement 1.. Aster formation activity of recombinant calmodulin-regulated spectrin-associated protein 3 (CAMSAP3).

(A) Schematic representation of the GFP-fused CAMSAP3 constructs used in this study. CH, calponin-homology domain; MBD, microtubule-binding domain; CC, coiled-coil domain; CKK, C-terminal domain common to CAMSAP1 and two other mammalian proteins, KIAA1078 and KIAA1543. (B) SDS-PAGE of GFP-fused CAMSAP3. (C) Negative stain electron microscopy (EM) micrographs of 10 µM tubulin polymerized with 1 µM CAMSAP3 after incubation on ice for 30 min and 37°C for 10 min.

Figure 5—figure supplement 2.

Figure 5—figure supplement 2.. Nucleation and aster formation activity of calmodulin-regulated spectrin-associated protein 2 (CAMSAP2).

(A) Negative stain images of 2 µM tubulin polymerized with 1 µM of CAMSAP2-FL after incubation at 37°C for 30 min in PEM. (B) Negative stain images of 10 µM tubulin polymerized with different concentrations of CAMSAP2-FL after incubation at 37°C for 30 min in PEM with 100 mM KCl. (C) Negative stain image of 10 µM tubulin polymerized with different concentrations of CAMSAP2-FL after incubation at 37°C for 30 min in PEM with 100 mM KCl.

Figure 5—figure supplement 3.

Figure 5—figure supplement 3.. The cryo-electron microscopy micrographs of 30 µM tubulin at different conditions.

No microtubule was observed. (Left) 30 µM tubulin preserved on ice for 30 min. (Middle left) Incubated for 1 min, (Middle right) 3 min, and (Right) 10 min at 37°C after 10 min on ice incubation. The bottom micrographs are closed-up views of red dotted areas indicated in the top micrographs.

Figure 6.

Figure 6.. Functional domain mapping of the microtubule nucleation and aster formation activity of calmodulin-regulated spectrin-associated protein 2 (CAMSAP2).

(A) Microtubule nucleation and aster formation activities of CAMSAP2 deletion constructs evaluated by the results of 10 µM tubulin with 1 µM CAMSAP2. The number of ‘+’ symbols indicates the strength of the activity (++++, strongest; +, weakest; ND, not detected). Size exclusion chromatography and SDS-PAGE of GFP fused constructs are available in Figure 6—figure supplement 1. CH, calponin-homology domain; MBD, microtubule-binding domain; CC, coiled-coil domain; CKK, C-terminal domain common to CAMSAP1 and two other mammalian proteins, KIAA1078 and KIAA1543. (B) Microtubule growth from CAMSAP2 condensates composed of full-length and deletion constructs. In vitro reconstitution was performed as described in Figure 4I. (C) Fluorescent intensity of CAMSAP2 condensates at 0 s. ND means that the fluorescent intensity of condensates could not be measured because GFP-CKK did not induce any condensates. Ordinary one-way ANOVA followed by Tukey’s multiple comparisons test. ****, p<0.0001. n=100 condensates from three independent preparations. (D) Negative stain EM images of polymerization by 10 µM tubulin with 1 µM GFP-CAMSAP2-FL or 1 µM CAMSAP2 mutants during 10 min of incubation at 37°C. The results for tubulin alone and GFP-CKK are available in Figure 6—figure supplement 3. Representative EM images are shown from at least three independent assays.

Figure 6—figure supplement 1.

Figure 6—figure supplement 1.. GFP fused calmodulin-regulated spectrin-associated protein 2 (CAMSAP2) constructs used in this study.

Size exclusion chromatography of GFP-CAMSAP2-FL (turquoise green), GFP-CC1-CKK (orange), GFP-CC3-CKK (yellow), GFP-CAMSAP2 ∆CC3 (magenta), GFP-CAMSAP2 ∆CKK (blue), and GFP-CKK (green). The peaks of each sample were subjected to SDS-PAGE.

Figure 6—figure supplement 2.

Figure 6—figure supplement 2.. Functional domain mapping of the microtubule nucleation and aster formation activity of calmodulin-regulated spectrin-associated protein 2 (CAMSAP2) deletion mutants.

(A) Schematic representation of the GFP fused CAMSAP2 deletion constructs. CH, calponin-homology domain; MBD, microtubule-binding domain; CC, coiled-coil domain; CKK, C-terminal domain common to CAMSAP1 and two other mammalian proteins, KIAA1078 and KIAA1543. (B) SDS-PAGE of the CAMSAP2 deletion constructs. (C) Negative stain electron microscopy (EM) micrographs of 10 µM tubulin polymerized with 1 µM of GFP-CC1-CKK-CT after incubation on ice for 30 min and 37°C for 10 min. (D) Negative stain EM micrographs of 10 µM tubulin polymerized with 1 µM of GFP-CC3-CKK-CT after incubation on ice for 30 min and 37°C for 10 min.

Figure 6—figure supplement 3.

Figure 6—figure supplement 3.. Functional domain mapping of calmodulin-regulated spectrin-associated protein 2 (CAMSAP2) analysed by negative stain electron microscopy (EM).

(A) Ten-micromolar tubulin was incubated alone (left), or with 1 µM GFP-CKK (middle), or with 1 µM GFP-CAMSAP2 ∆CKK (right) at 37°C for 10 min and analysed by negative stain EM. Oligomerization of tubulin was observed in these three conditions, albeit no microtubule formation was observed. (B) Ten-micromolar tubulin with 10 µM GFP-CKK was incubated at 37°C for 10 min and analysed by negative stain EM. Notably, the microtubules were fully decorated with the GFP-CKK.

Figure 7.

Figure 7.. Calmodulin-regulated spectrin-associated protein 2 (CAMSAP2) induces tubulin ring formation.

Representative electron microscopy (EM) images are shown from at least three independent assays. (A) Negative stain EM micrographs of 10 µM tubulin polymerization with 1 µM GFP-CC1-CKK at different time points. (B) Plots of the number of tubulin rings (orange) and that of microtubules (cyan) at different time points (mean ± SD, from 10 independent views).

Figure 8.

Figure 8.. Calmodulin-regulated spectrin-associated protein 2 (CAMSAP2) induced microtubule nucleation intermediates visualized by time-lapse cryo-electron microscopy (EM).

Representative EM images are shown from at least three independent assays. (A) Snapshots of growing microtubule intermediates at different time points. The contrast of the micrographs was adjusted to be slightly higher than the original (see Figure 8—figure supplement 2C for raw images). The black arrows indicate high-density condensed areas, red squares are segmented areas. Red arrows in the zoomed images indicate tubulin-rings or semi-rings, and blue arrows indicate microtubules or sheets. (B) Segmentation of the structural elements of micrographs from panel (A) (on ice, red box). Tubulin rings, red; intermediates between ring and sheet, orange; tubulin sheets, green. (C) Segmentation of the structural elements of the micrographs from (A) (60 s, red box). Tubulin rings, red; intermediates between ring and sheet, orange; tubulin sheets, green; microtubules, blue. (D) 2D classification of tubulin rings. Rings of different shapes and sizes were observed, including single rings, spiral rings, and double rings. (E) Comparison of 2D average of 28 tubulin rings (top) with the reported tubulin rings produced by the 30 tubulin ring consists of longitudinal contacts (EMD-7026). 30 tubulin ring (EMD-6347) was generated by cropping outer tubulin ring from the microtubule-KLP10A map (EMD-7026) using its model (PDB: 6b0c) as a guide. (F) Thirteen and sixteen (EMD-5196) PF microtubule (top) as examples for the ring diameters made by lateral contacts and the tubulin longitudinal tube (EMD-1131) (bottom) as an example for the ring diameter made by longitudinal contacts. Sixteen protofilaments are the thickest microtubule in EMDB. The scale bars indicate 10 nm. The cryo-Electron Tomography (ET) reconstruction of Cam2-asters is available in Figure 8—figure supplement 1 and Video 5.

Figure 8—figure supplement 1.

Figure 8—figure supplement 1.. Cryo-electron tomographic reconstruction of the growing Cam2-asters.

(A)-(E) Tomographic reconstruction of a growing Cam2-aster processed by SIRT. See also Video 5. (A) Rendered view of a growing Cam2-aster from representative 120 nm thick tomography slices overlaid with a 2.4 nm tomographic slice. Condensates of the Cam2-aster: yellow; microtubules: blue; tubulin sheets: green; tubulin rings: red. (B)-(E), Magnified views shown in panel (A).

Figure 8—figure supplement 2.

Figure 8—figure supplement 2.. Cryo-electron microscopy (Cryo-EM) visualization of Cam2-aster formation.

(A) Negative stain image of 10 µM tubulin on ice. (B) Cryo-EM image of 30 µM tubulin on ice without calmodulin-regulated spectrin-associated protein 2 (CAMSAP2). (C) Raw micrographs of snapshots of growing microtubule intermediates at different time points, displayed in Figure 8A. Segmented areas corresponding to Figure 8B and C are circled by red dotted square and enlarged. (D) Cryo-EM images of growing microtubules with CAMSAP2 CC1-CKK after 60 s of incubation at 37°C. The enlarged panel shows examples of partially curved sheet-like microtubule ends, which are characteristic of growing microtubules. Enlarged area shows tubulin sheets exhibiting partially curved sheet-like structures (E) Cryo-EM images of growing microtubules with CAMSAP2 CC1-CKK after 180 s of incubation at 37°C. The red arrows in panels C and D indicate high-density condensed areas.

Figure 8—figure supplement 3.

Figure 8—figure supplement 3.. Segmentation of the structural elements of the micrographs at 60 s.

Tubulin rings, red; intermediates between ring and sheet, orange; tubulin sheets, green; microtubules, blue.

Figure 9.

Figure 9.. Structural model of microtubule nucleation and Cam2-aster formation induced by calmodulin-regulated spectrin-associated protein 2 (CAMSAP2).

Structural model of tubulin nucleation, polymerization, and aster formation induced by CAMSAP2, as detailed in the main text. CAMSAP2 shifts the equilibrium to the right, as indicated by the arrow size.

Author response image 1.

Author response image 1.

Author response image 2.

Author response image 2.. GFP attached on the glass surface represents homogeneous signals, unlike GFP-CAMSAP2-FL.

References

    1. Akhmanova A, Steinmetz MO. Tracking the ends: a dynamic protein network controls the fate of microtubule tips. Nature Reviews. Molecular Cell Biology. 2008;9:309–322. doi: 10.1038/nrm2369. - DOI - PubMed
    1. Akhmanova A, Steinmetz MO. Control of microtubule organization and dynamics: two ends in the limelight. Nature Reviews. Molecular Cell Biology. 2015;16:711–726. doi: 10.1038/nrm4084. - DOI - PubMed
    1. Akhmanova A, Steinmetz MO. Microtubule minus-end regulation at a glance. Journal of Cell Science. 2019;132:jcs227850. doi: 10.1242/jcs.227850. - DOI - PubMed
    1. Al-Bassam J. Reconstituting dynamic microtubule polymerization regulation by TOG domain proteins. Methods in Enzymology. 2014;540:131–148. doi: 10.1016/B978-0-12-397924-7.00008-X. - DOI - PubMed
    1. Alushin GM, Lander GC, Kellogg EH, Zhang R, Baker D, Nogales E. High-resolution microtubule structures reveal the structural transitions in αβ-tubulin upon GTP hydrolysis. Cell. 2014;157:1117–1129. doi: 10.1016/j.cell.2014.03.053. - DOI - PMC - PubMed

Publication types

MeSH terms

Substances

Grants and funding

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

LinkOut - more resources