Clathrin heavy chain mediates TACC3 targeting to mitotic spindles to ensure spindle stability - PubMed (original) (raw)

Clathrin heavy chain mediates TACC3 targeting to mitotic spindles to ensure spindle stability

Chiou-Hong Lin et al. J Cell Biol. 2010.

Abstract

Mitotic spindles play essential roles in chromosome congression and segregation during mitosis. Aurora A regulates spindle assembly in part via phosphorylating human TACC3 on S558, which triggers TACC3 relocalization to mitotic spindles and stabilizes microtubules (MTs). In this study, we identified clathrin heavy chain (CHC) as an adaptor protein to recruit S558-phosphorylated TACC3 onto the spindle during mitosis for MT stabilization. CHC binds phospho-S558 TACC3 via its linker domain and first CHC repeat. CHC depletion or mutation on phospho-TACC3 binding abrogates TACC3 spindle relocalization. Depletion of either or both CHC and TACC3 yields similar defective phenotypes: loss of ch-TOG on spindles, disorganized spindles, and chromosome misalignment with comparable mitotic delay. Our findings elucidate the association between aurora A phosphorylation and spindle apparatus and demonstrate that regulation from aurora A is mediated by CHC in recruiting phospho-TACC3 and subsequently ch-TOG to mitotic spindles.

PubMed Disclaimer

Figures

Figure 1.

Figure 1.

CHC associates with phospho-S558 TACC3. (A) The SYPRO ruby gel shows CHC pulled down from Noc-treated HeLa cell extracts by recombinant GST-TACC3 522–577 fusion proteins phosphorylated by recombinant aurora A. CHC peptides detected by mass spectrometry are indicated. (B) Western blotting shows CHC pulled down by recombinant GST-TACC3 proteins phosphorylated by aurora A. Input represents the 5% amount of Noc-treated HeLa cell extracts subjected to the pull-down assays. Coomassie blue staining shows various GST fusion proteins used for each binding reaction. The phosphorylation levels of GST-TACC3 proteins in kinase reactions are shown by autoradiography. (C–E) Western blots show complex formation of endogenous TACC3 and CHC by immunoprecipitation (IP) with the indicated antibodies from mitotic HeLa cells synchronized by Noc (C), from Noc-synchronized cells with or without additional treatment of 2 µM VX-680 (D), or from Noc-synchronized cell lysates with or without treatment of λ-phosphatase (λPPase) before being subjected to immunoprecipitation (E). The phospho-T288 level correlates to aurora A kinase activity. Input represents the 5% amount of the indicated lysates for each immunoprecipitation. Black lines indicate that intervening lanes have been spliced out. (F and G) Representative images of HeLa cells in metaphase stained with DNA (blue) and antibodies against α-tubulin or CHC (green) and TACC3 (H-300) or phospho-TACC3 (pTACC3; red) are shown. Bars, 10 µm.

Figure 2.

Figure 2.

CHC is essential for TACC3 localization to the spindle. (A, B, and D) Metaphase images of HeLa cells treated with the indicated siRNAs for 72 h and stained with DNA (blue), TACC3 (H-300), phospho-TACC3 (pTACC3), aurora A, or phospho-T288 aurora A (red) and CHC or α-tubulin (green) as indicated. Bars, 10 µm. (C) Histogram shows TACC3 recruitment to the spindle in HeLa cells treated with the indicated siRNAs. Expression levels of the indicated proteins in siRNA-treated cells are shown. Relative spindle recruitment and spindle MT intensity are quantified as described in Materials and methods. Error bars indicate mean ± SD (_n_= 3; >20 mitotic cells scored per experiment). (E and F) Western blots show the levels of indicated proteins from mitotic extracts of cells treated with indicated siRNAs (E) and the level of indicated proteins from in vitro aster and MT-binding assays with mitotic extracts of cells treated with the indicated siRNAs (F). Black lines indicate that intervening lanes have been spliced out. Sup, supernatant.

Figure 3.

Figure 3.

Depletion of CHC and/or TACC3 causes comparable mitotic delay and chromosome alignment defect. (A) Western blotting shows the expression levels of CHC and TACC3 in HeLa cells treated with the indicated siRNAs. (B) Bar graph shows mitotic index of HeLa cells treated with the indicated siRNAs (n = 3; >500 cells scored per condition). (C) Box and whisker plot measuring time spent in mitosis of HeLa cells expressing GFP-H2B treated with the indicated siRNAs (>25 cells scored per condition; P < 0.0001). (D) Still images from

Video 1

of HeLa cells expressing GFP-H2B treated with the indicated siRNAs are shown. Arrows show chromosome misalignment. Bars, 10 µm. (E) Bar graph shows the percentage of mitotic HeLa cells with a metaphase-like plate and misaligned chromosome after treatment of the indicated siRNAs (n = 3; ∼100 cells per experiment). Error bars indicate mean ± SD.

Figure 4.

Figure 4.

Depletion of CHC and/or TACC3 renders a loss of ch-TOG spindle targeting and aberrant spindles. (A, D, and F) Images of HeLa cells treated with the indicated siRNAs and stained with DNA (blue), α-tubulin or CHC (green), and ch-TOG or hepatoma up-regulated protein (HURP; red) as indicated are shown. (B) Bar graph shows quantification of aberrant spindle morphology of cells as observed in A (n = 3; >100 cells per experiment). (C) Western blots show complex formation of endogenous CHC and ch-TOG by immunoprecipitation (IP) with the indicated antibodies from mitotic HeLa cells. Input represents the 5% amount of Noc-treated cell extracts subjected to immunoprecipitation. Black lines indicate that intervening lanes have been spliced out. (E) Bar graph shows the recruitment of ch-TOG to the spindle of cells observed in D (n = 3; >10 mitotic cells scored per experiment). (G) Images of MT repolymerization after cold shock of the indicated siRNA-treated cells. Cells were fixed and stained with α-tubulin (green) and DNA (blue). (H) Bar graph shows MT intensity and half-spindle length of cells observed in G (n = 3; >20 mitotic cells scored per experiment). Error bars indicate mean ± SD. Bars, 10 µm.

Figure 5.

Figure 5.

The CHC 331–542 region is important for TACC3 interaction and spindle recruitment. (A) A schematic presentation of the CHC domains and deletion mutants used in this study is shown. The interaction of each CHC deletion mutant with TACC3 is indicated. (B) Western blotting shows MBP-TACC3 522–577 recombinant proteins phosphorylated by aurora A and pulled down by the indicated GST fusion proteins. Input represents the 5% amount of MBP-TACC3 protein phosphorylated by aurora A used for each binding reaction and detected by anti-MBP antibody. The arrowhead and arrow indicate phosphorylated and unphosphorylated MBP-TACC3 522–577, respectively. Coomassie blue staining shows the GST fusion proteins used for each binding reaction. (C) Autoradiograph of in vitro 35S-labeled CHC WT or truncated proteins pulled down by GST-TACC3 522–577 proteins being phosphorylated by aurora A. Input and Coomassie blue staining represent 10% of the amount of in vitro–synthesized proteins and GST fusion proteins used for each binding reaction, respectively. (D) Images of HeLa cells treated with siCHC for 6 h and transfected with siCHC-resistant Flag-CHC WT or mutant for an additional 66 h and stained with DNA (blue), α-tubulin (green), and Flag-CHC (red). (E) Histogram shows the relative spindle recruitment of CHC WT or Δ(331–542) in siCHC-treated cells observed in D. The relative spindle MT intensity was normalized against the MT intensity obtained from siCHC and Δ(331–542)-transfected cells (>25 mitotic cells scored per construct). (F and G) Images of HeLa cells transfected with siCHC and the indicated siCHC-resistant Flag-CHC constructs as described in D were stained for Flag-CHC (green), TACC3 (H-300), or ch-TOG (red) and DNA (blue). (H and I) Bar graphs show mitotic index (H) or misaligned chromosomes (I) of HeLa cells transfected with the indicated siRNAs along with siCHC-resistant GFP-CHC constructs (n = 3; mitotic index, >500 cells scored; chromosome alignment defect, ∼100 cells scored). Error bars indicate mean ± SD. Bars, 10 µm.

References

    1. Barr A.R., Gergely F. 2007. Aurora-A: the maker and breaker of spindle poles. J. Cell Sci. 120:2987–2996 10.1242/jcs.013136 -DOI -PubMed
    1. Barr A.R., Gergely F. 2008. MCAK-independent functions of ch-Tog/XMAP215 in microtubule plus-end dynamics. Mol. Cell. Biol. 28:7199–7211 10.1128/MCB.01040-08 -DOI -PMC -PubMed
    1. Brittle A.L., Ohkura H. 2005. Centrosome maturation: Aurora lights the way to the poles. Curr. Biol. 15:R880–R882 10.1016/j.cub.2005.10.022 -DOI -PubMed
    1. Cassimeris L., Gard D., Tran P.T., Erickson H.P. 2001. XMAP215 is a long thin molecule that does not increase microtubule stiffness. J. Cell Sci. 114:3025–3033 -PubMed
    1. Gaglio T., Saredi A., Compton D.A. 1995. NuMA is required for the organization of microtubules into aster-like mitotic arrays. J. Cell Biol. 131:693–708 10.1083/jcb.131.3.693 -DOI -PMC -PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources