A small GTPase molecular switch regulates epigenetic centromere maintenance by stabilizing newly incorporated CENP-A

Journal name:
Nature Cell Biology
Year published:
Published online


Epigenetic mechanisms regulate genome activation in diverse events, including normal development and cancerous transformation. Centromeres are epigenetically designated chromosomal regions that maintain genomic stability by directing chromosome segregation during cell division. The histone H3 variant CENP-A resides specifically at centromeres, is fundamental to centromere function and is thought to act as the epigenetic mark defining centromere loci. Mechanisms directing assembly of CENP-A nucleosomes have recently emerged, but how CENP-A is maintained after assembly is unknown. Here, we show that a small GTPase switch functions to maintain newly assembled CENP-A nucleosomes. Using functional proteomics, we found that MgcRacGAP (a Rho family GTPase activating protein) interacts with the CENP-A licensing factor HsKNL2. High-resolution live-cell imaging assays, designed in this study, demonstrated that MgcRacGAP, the Rho family guanine nucleotide exchange factor (GEF) Ect2, and the small GTPases Cdc42 and Rac, are required for stability of newly incorporated CENP-A at centromeres. Thus, a small GTPase switch ensures epigenetic centromere maintenance after loading of new CENP-A.

At a glance


  1. MgcRacGAP is required for CENP-A protein localization to centromeres.
    Figure 1: MgcRacGAP is required for CENP-A protein localization to centromeres.

    (a) Model of the licensing (1) and loading (2) steps of CENP-A into centromere chromatin shows the hypothetical replacement of H3 nucleosomes with CENP-A (see text for details). (b) Representative immunofluorescence images of HeLa cells transfected with shRNA specific to MgcRacGAP and expressing CENP-A–YFP (middle) and stained with antibodies specific to HsKNL2. Top image is merged from bottom two images. Depleted cells were identified by co-transfection with RFP–Histone H2B (red in merge); an untransfected control cell is also indicated. HsKNL2 localization (bottom panel) is missing from the control because HsKNL2 is normally lost after the end of G1. Thus, HsKNL2 localization is normal in both cases. Scale bar, 10 μm. (c) Intensity of CENP-A–YFP at centromeres in cells expressing CENP-A–YFP and transfected with the indicated shRNA plasmids, as assessed by high-resolution imaging of cells. Data are means ± s.e.m., n; number of cells analysed. (d) Western blot of lysates from control cells (diluted to the indicated percentages), compared with lysates from cells transfected with MgcRacGAP shRNA (right). The indicated antibodies were used for blotting. All depletions were confirmed by qPCR (Supplementary Information, Table S6).

  2. MgcRacGAP localizes to centromeres transiently at the end of CENP-A loading.
    Figure 2: MgcRacGAP localizes to centromeres transiently at the end of CENP-A loading.

    (a) Representative fluorescence microscopy images of endogenous MgcRacGAP and ACA (top; stained with indicated antibodies) and exogenous MgcRacGAP and Mis18 (MgcRacGAP–mCherry and Mis18–GFP; bottom panels) in HeLa cells in G1 phase. Arrows indicate co-localization of MgcRacGAP and centromeres. Images on the right are merged from images on the left, and insets are zooms of indicated nuclear regions; tubulin staining is overlaid in the top right panel in grayscale. Scale bar in nuclear insets, 2 μm and in merged image, 10 μm. (b) Cells expressing MgcRacGAP–mCherry and CENP-A–YFP were imaged by time-lapse microscopy. Representative images are shown of MgcRacGAP–mCherry (left) and CENP-A–YFP localization (middle) at indicated times after initiation of G1 phase. Insets are ×2 zoom of indicated nuclear regions. Scale bar, 5 μm. (c) Quantification of CENP-A–YFP intensity (A.U.; arbitrary units) at centromeres and number of MgcRacGAP–mCherry spots that co-localize with centromeres, from an experiment performed as in b. (d) Schematic representation of CENP-A (green), HsKNL2 (blue) and MgcRacGAP (red) localization to centromeres with respect to the cell cycle. M is mitosis and S is S phase.

  3. MgcRacGAP is required specifically to stabilize newly incorporated CENP-A.
    Figure 3: MgcRacGAP is required specifically to stabilize newly incorporated CENP-A.

    (a) Newly deposited CENP-A is specifically lost in the absence of MgcRacGAP. Cells stably expressing SNAP-labelled CENP-A were pulse-labelled with TMR. After pulse-labelling, total CENP-A was labelled by immunostaining (green; middle). Images indicate cells after treatment with (bottom) or without (control; top) MgcRacGAP shRNA. The average intensity of signal at centromeres in the cell, normalized with respect to the control cell, is reported in the lower left corner. Colour overlay and a colour scale of the relative ratio (right) indicate that MgcRacGAP is required for maintenance of newly incorporated centromere CENP-A. Scale bar, 5 μm. (b) Quantification of SNAP-tag-labelled CENP-A and total CENP-A in cells treated with MgcRacGAP shRNA, compared with control cells from experiment performed as in a. Levels of SNAP-labelled CENP-A (old, red) were relatively unchanged (P = 0.5), but total CENP-A (green) was reduced to less than 50% (P < 0.001) after MgcRacGAP depletion, compared with controls. This indicates that the CENP-A lost was not at centromeres when the SNAP pulse label was administered in the previous cell cycle and was therefore newly incorporated CENP-A protein. Data are means ± s.e.m.

  4. GAP-inactive MgcRagGAP mutant localizes persistently to centromeres.
    Figure 4: GAP-inactive MgcRagGAP mutant localizes persistently to centromeres.

    (a) Fluorescence microscopy images of a HeLa cell expressing GAP-inactive MgcRacGAP–mCherry and CENP-A–YFP. Scale bar, 10 μm. (b) Quantification of CENP-A–YFP intensity at centromeres and number of GAP-inactive MgcRacGAP–mCherry spots that co-localize with centromeres, as assessed from an automated time-lapse live-cell analysis. (c) The number of MgcRacGAP spots that co-localize with centromeres in cells expressing GAP-inactive MgcRacGAP–mCherry or wild-type MgcRacGAP–mCherry was compared by live-cell imaging. Co-localized spots were identified by visual inspection of dual-colour images. Blue, number of co-localizing spots per analysed cell; red, mean ± s.d. Control n = 22 cells; GAP dead n = 33 cells (d) Live-cell imaging was used to quantify time of co-localization in cells expressing CENP-A–YFP and either GAP-inactive MgcRacGAP–mCherry or wild-type MgcRacGAP–mCherry. Each row corresponds to a centromere that was tracked through time (horizontal axis). Black dots indicate co-localization of MgcRacGAP–mCherry with the centromere track.

  5. Automated analysis of CENP-A levels following shRNA depletion of various target proteins reveals differential defects in epigenetic regulation of centromeres.
    Figure 5: Automated analysis of CENP-A levels following shRNA depletion of various target proteins reveals differential defects in epigenetic regulation of centromeres.

    Representative images of cells at ×60 magnification from cells treated with indicated shRNA. Cells were stained with DAPI and the nuclei were segmented (Supplementary Information, Fig. S3). Images of DAPI (red) and CENP-A–YFP (green) localization are overlaid in the left column. The column labelled 'Auto' shows the CENP-A-YFP channel rescaled from 12 bits to 8 bits (required for display purposes) using an auto-scale where the highest pixel value is assigned 255 and the lowest 0 on the 8 bit scale. The column labelled 'Min' shows the same images as the Auto column, but is scaled from 12 bits to 8 bits using the dynamic range of the dimmest cell in the field of view (indicated by the white box in the Auto image). The resulting image reveals cells that have reduced CENP-A signal (affected cells) and saturates cells with 'normal' signal (unaffected cells). Comparing the Auto and Min columns gives a quick view of the level of CENP-A lost in various treatments; conditions resulting in greater loss of CENP-A have more bright cells in the Min column. In the last column, the Auto column has been subtracted from the Min column and pseudo-coloured to facilitate visualization. Redder colours indicate a greater difference between the Min and Auto columns. Therefore, fields with more red colours represent a greater effect on CENP-A levels. Numbers on the right of each image give the mean ± s.e.m. of CENP-A intensity per cell. n; number of cells measured. Scale bar, 20 μm.

  6. Cdc42 localizes to centromeres and functions to maintain CENP-A levels independent of polymeric actin.
    Figure 6: Cdc42 localizes to centromeres and functions to maintain CENP-A levels independent of polymeric actin.

    (a) Representative fluorescence microscopy images of cells stained with antibodies against the indicated GTPases (left). Cells were expressing CENP-A–YFP (as shown in the middle images). Right: merge of images on the left. Co-localization of endogenous Cdc42 and CENP-A–YFP is indicated in the insets. All images are projections of 4–8 optical sections from deconvolved image stacks. Scale bar, 5 μm. (b) Representative fluorescence microscopy images of cells expressing CENP-A–YFP and stained with rhodamine-phalloidin to label polymeric actin. Cells were treated as indicated. Right: merge of images on the left, with DAPI used to stain nuclei. Scale bar, 20 μm. (c) Quantification of CENP-A–YFP intensity at centromeres in cells treated as indicated in b, normalized to cells treated with DMSO. Data are means ± s.e.m.; n, number of cells analysed.

  7. A small GTPase switch maintains newly incorporated CENP-A after loading.
    Figure 7: A small GTPase switch maintains newly incorporated CENP-A after loading.

    During G1, the centromere is specified in a three-step process. After licensing (1) and loading (2; see also Fig. 1), MgcRacGAP and ECT2 cycle Cdc42 GTPase activity to modify newly incorporated CENP-A nucleosomes to make them molecularly identical to pre-existing CENP-A nucleosomes (Step 3).


  1. Goldberg, A. D., Allis, C. D. & Bernstein, E. Epigenetics: a landscape takes shape. Cell 128, 635638 (2007).
  2. Karpen, G. H. & Allshire, R. C. The case for epigenetic effects on centromere identity and function. Trends Genet. 13, 489496 (1997).
  3. Palmer, D. K., O'Day, K., Trong, H. L., Charbonneau, H. & Margolis, R. L. Purification of the centromere-specific protein CENP-A and demonstration that it is a distinctive histone. Proc. Natl Acad. Sci. USA 88, 37343738 (1991).
  4. Choo, K. H. Centromerization. Trends Cell Biol. 10, 182188 (2000).
  5. Bernad, R., Sanchez, P. & Losada, A. Epigenetic specification of centromeres by CENP-A. Exp. Cell Res. 315, 32333241 (2009).
  6. Sullivan, K. F., Hechenberger, M. & Masri, K. Human CENP-A contains a histone H3 related histone fold domain that is required for targeting to the centromere. J. Cell Biol. 127, 581592 (1994).
  7. Cleveland, D. W., Mao, Y. & Sullivan, K. F. Centromeres and kinetochores: from epigenetics to mitotic checkpoint signaling. Cell 112, 407421 (2003).
  8. Schuh, M., Lehner, C. F. & Heidmann, S. Incorporation of Drosophila CID/CENP-A and CENP-C into centromeres during early embryonic anaphase. Curr. Biol. 17, 237243 (2007).
  9. Jansen, L. E., Black, B. E., Foltz, D. R. & Cleveland, D. W. Propagation of centromeric chromatin requires exit from mitosis. J. Cell Biol. 176, 795805 (2007).
  10. Hayashi, T. et al. Mis16 and Mis18 are required for CENP-A loading and histone deacetylation at centromeres. Cell 118, 715729 (2004).
  11. Fujita, Y. et al. Priming of centromere for CENP-A recruitment by human hMis18α, hMis18β, and M18BP1. Dev. Cell 12, 1730 (2007).
  12. Maddox, P. S., Hyndman, F., Monen, J., Oegema, K. & Desai, A. Functional genomics identifies a Myb domain-containing protein family required for assembly of CENP-A chromatin. J. Cell Biol. 176, 757763 (2007).
  13. Foltz, D. R. et al. Centromere-specific assembly of CENP-A nucleosomes is mediated by HJURP. Cell 137, 472484 (2009).
  14. Dunleavy, E. M. et al. HJURP is a cell-cycle-dependent maintenance and deposition factor of CENP-A at centromeres. Cell 137, 485497 (2009).
  15. Silva, M. C. & Jansen, L. E. At the right place at the right time: novel CENP-A binding proteins shed light on centromere assembly. Chromosoma 118, 567574 (2009).
  16. Perpelescu, M., Nozaki, N., Obuse, C., Yang, H. & Yoda, K. Active establishment of centromeric CENP-A chromatin by RSF complex. J. Cell Biol. 185, 397407 (2009).
  17. Etienne-Manneville, S. & Hall, A. Rho GTPases in cell biology. Nature 420, 629635 (2002).
  18. Izuta, H. et al. Comprehensive analysis of the ICEN (interphase centromere complex) components enriched in the CENP-A chromatin of human cells. Genes Cells 11, 673684 (2006).
  19. Kawashima, T. et al. MgcRacGAP is involved in the control of growth and differentiation of hematopoietic cells. Blood 96, 21162124 (2000).
  20. Minoshima, Y. et al. Phosphorylation by aurora B converts MgcRacGAP to a RhoGAP during cytokinesis. Dev. Cell 4, 549560 (2003).
  21. Toure, A. et al. MgcRacGAP, a new human GTPase-activating protein for Rac and Cdc42 similar to Drosophila rotundRacGAP gene product, is expressed in male germ cells. J. Biol. Chem. 273, 60196023 (1998).
  22. Yuce, O., Piekny, A. & Glotzer, M. An ECT2-centralspindlin complex regulates the localization and function of RhoA. J. Cell Biol. 170, 571582 (2005).
  23. Canman, J. C. et al. Inhibition of Rac by the GAP activity of centralspindlin is essential for cytokinesis. Science 322, 15431546 (2008).
  24. Miller, A. L. & Bement, W. M. Regulation of cytokinesis by Rho GTPase flux. Nat. Cell Biol. 11, 7177 (2009).
  25. Hirose, K., Kawashima, T., Iwamoto, I., Nosaka, T. & Kitamura, T. MgcRacGAP is involved in cytokinesis through associating with mitotic spindle and midbody. J. Biol. Chem. 276, 58215828 (2001).
  26. Jantsch-Plunger, V. et al. CYK-4: A Rho family gtpase activating protein (GAP) required for central spindle formation and cytokinesis. J. Cell Biol. 149, 13911404 (2000).
  27. Pavicic-Kaltenbrunner, V., Mishima, M. & Glotzer, M. Cooperative assembly of CYK-4/MgcRacGAP and ZEN-4/MKLP1 to form the centralspindlin complex. Mol. Biol. Cell 18, 49925003 (2007).
  28. Maddox, A. S. & Oegema, K. Closing the GAP: a role for a RhoA GAP in cytokinesis. Mol. Cell 11, 846848 (2003).
  29. Black, B. E. et al. Centromere identity maintained by nucleosomes assembled with histone H3 containing the CENP-A targeting domain. Mol. Cell 25, 309322 (2007).
  30. Foltz, D. R. et al. The human CENP-A centromeric nucleosome-associated complex. Nat. Cell Biol. 8, 458469 (2006).
  31. von Dassow, G., Verbrugghe, K. J., Miller, A. L., Sider, J. R. & Bement, W. M. Action at a distance during cytokinesis. J. Cell Biol. 187, 831845 (2009).
  32. Yasuda, S. et al. Cdc42 and mDia3 regulate microtubule attachment to kinetochores. Nature 428, 767771 (2004).
  33. Oceguera-Yanez, F. et al. Ect2 and MgcRacGAP regulate the activation and function of Cdc42 in mitosis. J. Cell Biol. 168, 221232 (2005).
  34. Sanders, L. C., Matsumura, F., Bokoch, G. M. & de Lanerolle, P. Inhibition of myosin light chain kinase by p21-activated kinase. Science 283, 20832085 (1999).
  35. Rando, O. J., Zhao, K. & Crabtree, G. R. Searching for a function for nuclear actin. Trends Cell Biol. 10, 9297 (2000).
  36. Glotzer, M. The molecular requirements for cytokinesis. Science 307, 17351739 (2005).
  37. Maddox, A. S. & Oegema, K. Deconstructing cytokinesis. Nat. Cell Biol. 5, 773776 (2003).
  38. Field, C., Li, R. & Oegema, K. Cytokinesis in eukaryotes: a mechanistic comparison. Curr. Opin. Cell Biol. 11, 6880 (1999).
  39. Chen, M. & Shen, X. Nuclear actin and actin-related proteins in chromatin dynamics. Curr. Opin. Cell Biol. 19, 326330 (2007).
  40. Tomonaga, T. et al. Overexpression and mistargeting of centromere protein-A in human primary colorectal cancer. Cancer Res. 63, 35113516 (2003).
  41. Collins, K. A., Furuyama, S. & Biggins, S. Proteolysis contributes to the exclusive centromere localization of the yeast Cse4/CENP-A histone H3 variant. Curr. Biol. 14, 19681972 (2004).
  42. Ye, J., Zhao, J., Hoffmann-Rohrer, U. & Grummt, I. Nuclear myosin I acts in concert with polymeric actin to drive RNA polymerase I transcription. Genes Dev. 22, 322330 (2008).
  43. Theriot, J. A., Mitchison, T. J., Tilney, L. G. & Portnoy, D. A. The rate of actin-based motility of intracellular Listeria monocytogenes equals the rate of actin polymerization. Nature 357, 257260 (1992).
  44. Stelzer, E. H., Wacker, I. & De Mey, J. R. Confocal fluorescence microscopy in modern cell biology. Semin. Cell Biol. 2, 145152 (1991).
  45. Otsu, N. A threshold selection method from gray-level histograms. IEEE Transactions Syst. Man Cybern. 9, 6266 (1979).
  46. Lindeberg, T. & Eklundh, J. O. Scale detection and region extraction from a scale-space primal sketch. Proc. 3rd International Conference on Computer Vision (Osaka, Japan) 416426 (1990).
  47. Dorn, J. F., Danuser, G. & Yang, G. Computational processing and analysis of dynamic fluorescence image data. Methods Cell Biol. 85, 497538 (2008).
  48. Jaqaman, K. et al. Kinetochore alignment within the metaphase plate is regulated by centromere stiffness and microtubule depolymerases. J. Cell Biol. 188, 665679 (2010).
  49. Dorn, J. F. et al. Yeast kinetochore microtubule dynamics analyzed by high-resolution three-dimensional microscopy. Biophys. J. 89, 28352854 (2005).
  50. Carroll, C. W., Silva, M. C., Godek, K. M., Jansen, L. E. & Straight, A. F. Centromere assembly requires the direct recognition of CENP-A nucleosomes by CENP-N. Nat. Cell Biol. 11, 896902 (2009).
  51. Cheeseman, I. M. & Desai, A. A combined approach for the localization and tandem affinity purification of protein complexes from metazoans. Sci STKE 2005, pl1 (2005).
  52. Cheeseman, I. M. et al. A conserved protein network controls assembly of the outer kinetochore and its ability to sustain tension. Genes Dev. 18, 22552268 (2004).

Download references

Author information

  1. These authors contributed equally to this work.

    • Anaïck Lagana &
    • Jonas F. Dorn


  1. Institute for Research in Immunology and Cancer (IRIC), Université de Montréal P.O. Box 6128, Station Centre-Ville Montréal QC, H3C 3J7 Canada.

    • Anaïck Lagana,
    • Jonas F. Dorn,
    • Valérie De Rop,
    • Anne-Marie Ladouceur,
    • Amy S. Maddox &
    • Paul S. Maddox
  2. Department of Pathology and Cell Biology, Université de Montréal P.O. Box 6128, Station Centre-Ville Montréal QC, H3C 3J7 Canada.

    • Amy S. Maddox &
    • Paul S. Maddox


A.L. conducted all immunoprecipitation and mass spectrometry experiments. A.L. and J.F.D. performed shRNA experiments and analysis, respectively. J.F.D. performed live-cell experiments. A.L. evaluated all shRNA experiments (qPCR, western blot). V.D.R. performed the small GTPases localization experiments. All authors performed essential tasks in generating figures. P.S.M., A.L., J.D., A.-M.L., and A.S.M. conceived the experiments. P.S.M. wrote the manuscript, assisted by all authors, in particular A.S.M. and J.F.D.

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to:

Author details

Supplementary information

PDF files

  1. Supplementary Information (945K)

    Supplementary Information


  1. Supplementary Movie 1 (2M)

    Supplementary Information

  2. Supplementary Movie 2 (235K)

    Supplementary Information

Additional data