Research ArticleMOLECULAR BIOLOGY

A helical inner scaffold provides a structural basis for centriole cohesion

See allHide authors and affiliations

Science Advances  14 Feb 2020:
Vol. 6, no. 7, eaaz4137
DOI: 10.1126/sciadv.aaz4137
  • Fig. 1 A circular scaffold inside the centriole resists deformation forces to help hold MTTs together.

    (A) Cryo–electron tomogram of a P. tetraurelia centriole showing the entire structure in longitudinal view (left) and ninefold symmetrized cross sections from two regions along its length (right). Orange, inner scaffold; turquoise, A-C linker. (B and C) Comparison of four species: (B) position along the centriole of the different MTT-attached structures and (C) percentage of the centriole covered by each structure. LECA, last eukaryotic common ancestor; Orange, inner scaffold; turquoise, A-C linker; gray, microtubule wall. Each dot in (C) is a different centriole. Error bars denote SD. N = 35 (P. tetraurelia), N = 6 (C. reinhardtii), N = 11 (N. gruberi), and N = 12 (human). (D and E) Cryo-ET cross sections through compressed P. tetraurelia centrioles (left) with corresponding schematic representations (right). Orange arrowheads indicate the inner scaffold attached to the MTTs. (F) Percentage of total P. tetraurelia centrioles (compressed and uncompressed) that are intact (black), have partially detached MTTs (white), or are broken (gray) in the central (N = 90 centrioles) and proximal (N = 72 centrioles) regions. (G) Schematic top views of a centriole at different cross sections showing the change in MTT orientation. Orange, inner scaffold; turquoise, A-C linker; gray, cartwheel. (H and I) MTT twist angle along the centriole length in P. tetraurelia (H) and in C. reinhardtii (I). Note that the A-C linker and the inner scaffold overlap in the region spanning 30 to 40%. Scale bars, 100 nm. n.s., not significant. *0.01 < P < 0.05; **0.001 < P < 0.01; ***0.0001 < P < 0.001; **** P < 0.0001.

  • Fig. 2 Subtomogram averaging from uncompressed P. tetraurelia and C. reinhardtii centrioles reveals how the inner scaffold links adjacent MTTs.

    Cryo-ET reconstructions of the inner scaffold and MTT structure from (A and B) P. tetraurelia and (C and D) C. reinhardtii. (A and C) Top views showing the centrioles in cross section. Purple, MTT protofilaments; yellow/orange, inner scaffold including densities that correspond to the C. reinhardtii Y-shaped linker; pink/red, A-microtubule MIPs (microtubule inner proteins); lavender, structures attached to the A-microtubule seam (between A9 and A10); turquoise, B-microtubule MIPs; green, structures attached to the C-microtubule. (B and D) Lateral views from the centriole interior showing adjacent MTTs connected by the inner scaffold; color code described above. Scale bars, 20 nm.

  • Fig. 3 The inner scaffold forms a dense helical lattice.

    (A and E) Three-dimensional (3D) views of the ninefold symmetric central regions from (A) P. tetraurelia and (E) C. reinhardtii centrioles. Purple, MTTs; orange, inner scaffold. (B and F) Unrolled structures of the (B) P. tetraurelia and (F) C. reinhardtii centrioles. Pink, multiple repeats of one helix. In P. tetraurelia, the complete lattice of the inner scaffold is a two-start helix, with a periodicity of 17 nm that allows a second helix (orange) to fit within the scaffold. In C. reinhardtii, the inner scaffold lattice is a three-start helix, with a periodicity of 49.5 nm that allows two other helices (orange) to fit within the scaffold. (C, D, G, and H) Helical parameters of the (C and D) P. tetraurelia and (G and H) C. reinhardtii inner scaffolds shown in (C and G) 3D and (D and H) unrolled schematic views, colored as in (B) and (F). (I and J) Longitudinal sections through 3D models of the full-length (I) P. tetraurelia and (J) C. reinhardtii central regions. Scale bars, 50 nm.

  • Fig. 4 Molecular composition of the inner scaffold revealed by U-ExM.

    (A to C) Representative confocal images of in situ human centrosomes in U-ExM stained for tubulin (magenta) together with (A) HsSAS-6 (green), (B) FAM161A (green), or (C) CP110 (green). Note the mature centrioles (M) with their respective procentrioles (P). Insets show procentrioles from the dashed boxes. Scale bar, 200 nm. (D to G) Representative confocal images of in situ mature centrioles in U-ExM (longitudinal view) stained for tubulin (magenta) together with (D) POC1B (cyan), (E) FAM161A (green), (F) POC5 (yellow), or (G) Centrin-2 (gray). Dt, distal tip; C, central core; P, proximal region. Scale bar, 100 nm. (H) Position of each stained protein along the centriole with their respective percentages of centriole coverage. Error bars denote SD. Averages and SDs are as follows: POC1B, 67 ± 13% (N = 56); FAM161A, 64 ± 11% (N = 58); POC5, 65 ± 1% (N = 57); and Centrin-2, 79 ± 11% (N = 59). (I) Representative top view confocal images of human centrioles in U-ExM stained as described above. Scale bar, 100 nm. (J) Ninefold symmetrized images of the indicated conditions from (I). For comparison, simulated U-ExM signal for the inner scaffold (turquoise) and tubulin (magenta) was generated from the cryo-EM structure. (K) Average positions of fluorescent molecular labels for tubulin (magenta), POC1B (cyan), FAM161A (green), POC5 (yellow), and Centrin-2 (gray) superimposed on the centriole structure, seen in top view. The position of the inner scaffold is depicted in dashed black lines, and the width of its simulated U-ExM signal is in gray (related to fig. S10, D and E). Gray box, zoom in of (K). Scale bar, 20 nm.

  • Fig. 5 FAM161A, POC5, POC1B, and Centrin form a complex that binds microtubules.

    U2OS cells transfected with (A) mCherry-FAM161A, (B) GFP-POC5, (C) mCherry-FAM161A and GFP-POC5, (D) GFP-POC1B, (E) mCherry-FAM161A and GFP-POC1B, (F) GFP–Centrin-2, (G) mCherry-FAM161A and GFP–Centrin-2, (H) mCherry-POC5 and GFP–Centrin-2, and (I) mCherry-FAM161A and both GFP–Centrin-2 and mCherry-POC5. Note that POC1B and POC5, but not Centrin-2, are relocalized to microtubules when cotransfected with FAM161A. GFP–Centrin-2 is relocalized to FAM161A-decorated microtubules only in the presence of POC5, indicating that the three proteins form a complex, with POC5 mediating the interaction between Centrin-2 and FAM161A. Scale bar, 10 μm. Percentage of GFP/mCherry-positive cells with proteins localized to cytosol (Cyt.) or microtubules (MT) for each condition. Averages and SDs are as follows: (A) Cyt.: 11 ± 1.2; MT: 89 ± 1.2, (B) Cyt.: 100; MT: 0, (C) Cyt.: 7 ± 1.0; MT: 93 ± 1.0, (D) Cyt.: 100; MT: 0, (E) Cyt.: 7 ± 1.6; MT: 93 ± 1.6, (F) Cyt.: 100; MT: 0, (G) Cyt.: 100; MT: 0, (H) Cyt.: 100; MT: 0, and (I) Cyt.: 1 ± 0.7; MT: 99 ± 0.7 (for each protein). N = ~100 cells per condition from three independent experiments. (J) Coimmunoprecipitation assay in human embryonic kidney (HEK) cells cotransfected with GFP-FAM161A and mCherry-POC5 (top), GFP–Centrin-2 and mCherry-FAM161A as a negative control (middle), or GFP-POC5 and mCherry-POC1B (bottom). I, input; U, unbound; GFP-IP, immunoprecipitated material. Samples were run on a 4 to 20% gradient gel and transferred to polyvinylidene difluoride membranes before immunoblotting against GFP and mCherry. Data are from three independent experiments. (K) Full cryo-EM structural model of the centriole, with the inner scaffold covering 70% of the total length. The U-ExM localization of each inner scaffold protein is superimposed on the model as in Fig. 4K. Scale bar, 20 nm. (L) Schematic of the molecular interactions within the complex formed by FAM161A, POC1B, POC5, and Centrin-2, overlaid on the structural model of the inner scaffold seen in top view. Asterisks indicate coimmunoprecipitation results from this study.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/6/7/eaaz4137/DC1

    Fig. S1. Visualization of the connections between MTTs in cryo-tomograms.

    Fig. S2. MTT to MTD transitions.

    Fig. S3. The inner scaffold maintains centriole cohesion under compressive forces.

    Fig. S4. Reconstruction resolutions and computational pipeline for reconstructing the complete central region.

    Fig. S5. Assessing the correctness of the structure: Raw images versus averages.

    Fig. S6. Periodicity of the inner scaffold stem in P. tetraurelia, C. reinhardtii, N. gruberi, and humans.

    Fig. S7. Periodicity, helicity, and de novo subtomogram averaging.

    Fig. S8. Optimized U-ExM protocol in human cells reveals components of the inner scaffold.

    Fig. S9. POC1B, FAM161A, POC5, and Centrin-2 colocalize with the inner scaffold structure as simulated U-ExM fluorescence.

    Fig. S10. 3D localization of the inner core proteins and extended localization of GFP-POC1B at the proximal region of the centriole upon overexpression.

    Movie S1. Cryo-tomogram of a P. tetraurelia centriole ex vivo.

    Movie S2. Cryo-tomogram of a C. reinhardtii centriole in situ.

    Movie S3. 3D representation of the P. tetraurelia central core highlighting the helical pattern of the inner scaffold.

    Movie S4. 3D representation of the C. reinhardtii central core highlighting the helical pattern of the inner scaffold.

    Movie S5. 3D rendering of the full central core architecture from P. tetraurelia.

    Movie S6. 3D rendering of the full central core architecture from C. reinhardtii.

    Data file S1. The complete central core reconstruction from P. tetraurelia.

    Data file S2. The complete central core reconstruction from C. reinhardtii.

    Reference (40)

  • Supplementary Materials

    The PDF file includes:

    • Fig. S1. Visualization of the connections between MTTs in cryo-tomograms.
    • Fig. S2. MTT to MTD transitions.
    • Fig. S3. The inner scaffold maintains centriole cohesion under compressive forces.
    • Fig. S4. Reconstruction resolutions and computational pipeline for reconstructing the complete central region.
    • Fig. S5. Assessing the correctness of the structure: Raw images versus averages.
    • Fig. S6. Periodicity of the inner scaffold stem in P. tetraurelia, C. reinhardtii, N. gruberi, and humans.
    • Fig. S7. Periodicity, helicity, and de novo subtomogram averaging.
    • Fig. S8. Optimized U-ExM protocol in human cells reveals components of the inner scaffold.
    • Fig. S9. POC1B, FAM161A, POC5, and Centrin-2 colocalize with the inner scaffold structure as simulated U-ExM fluorescence.
    • Fig. S10. 3D localization of the inner core proteins and extended localization of GFP-POC1B at the proximal region of the centriole upon overexpression.
    • Legends for movies S1 to S6
    • Legends for data files S1 and S2
    • Reference (40)

    Download PDF

    Other Supplementary Material for this manuscript includes the following:

    • Movie S1 (.avi format). Cryo-tomogram of a P. tetraurelia centriole ex vivo.
    • Movie S2 (.avi format). Cryo-tomogram of a C. reinhardtii centriole in situ.
    • Movie S3 (.mov format). 3D representation of the P. tetraurelia central core highlighting the helical pattern of the inner scaffold.
    • Movie S4 (.mov format). 3D representation of the C. reinhardtii central core highlighting the helical pattern of the inner scaffold.
    • Movie S5 (.mov format). 3D rendering of the full central core architecture from P. tetraurelia.
    • Movie S6 (.mov format). 3D rendering of the full central core architecture from C. reinhardtii.
    • Data file S1 (.tif format). The complete central core reconstruction from P. tetraurelia.
    • Data file S2 (.tif format). The complete central core reconstruction from C. reinhardtii.

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article