During asymmetric division, fate determinants at the cell cortex segregate unequally into the two daughter cells. It has recently been shown that Sara (Smad anchor for receptor activation) signalling endosomes in the cytoplasm also segregate asymmetrically during asymmetric division1,2. Biased dispatch of Sara endosomes mediates asymmetric Notch/Delta signalling during the asymmetric division of sensory organ precursors in Drosophila1. In flies, this has been generalized to stem cells in the gut3 and the central nervous system1, and, in zebrafish, to neural precursors of the spinal cord4. However, the mechanism of asymmetric endosome segregation is not understood. Here we show that the plus-end kinesin motor Klp98A targets Sara endosomes to the central spindle, where they move bidirectionally on an antiparallel array of microtubules. The microtubule depolymerizing kinesin Klp10A and its antagonist Patronin generate central spindle asymmetry. This asymmetric spindle, in turn, polarizes endosome motility, ultimately causing asymmetric endosome dispatch into one daughter cell. We demonstrate this mechanism by inverting the polarity of the central spindle by polar targeting of Patronin using nanobodies (single-domain antibodies). This spindle inversion targets the endosomes to the wrong cell. Our data uncover the molecular and physical mechanism by which organelles localized away from the cellular cortex can be dispatched asymmetrically during asymmetric division.
This is a preview of subscription content, access via your institution
Open Access articles citing this article.
Polarized branched Actin modulates cortical mechanics to produce unequal-size daughters during asymmetric division
Nature Cell Biology Open Access 06 February 2023
Nature Communications Open Access 20 November 2019
BMC Molecular and Cell Biology Open Access 17 April 2019
Subscribe to this journal
Receive 51 print issues and online access
$199.00 per year
only $3.90 per issue
Rent or buy this article
Prices vary by article type
Prices may be subject to local taxes which are calculated during checkout
Coumailleau, F., Furthauer, M., Knoblich, J. A. & Gonzalez-Gaitan, M. Directional Delta and Notch trafficking in Sara endosomes during asymmetric cell division. Nature 458, 1051–1055 (2009)
Loubéry, S. et al. Uninflatable and Notch control the targeting of Sara endosomes during asymmetric division. Curr. Biol. 24, 2142–2148 (2014)
Montagne, C. & Gonzalez-Gaitan, M. Sara endosomes and the asymmetric division of intestinal stem cells. Development 141, 2014–2023 (2014)
Kressmann, S., Campos, C., Castanon, I., Furthauer, M. & Gonzalez-Gaitan, M. Directional Notch trafficking in Sara endosomes during asymmetric cell division in the spinal cord. Nature Cell Biol. 17, 333–339 (2015)
Hoepfner, S. et al. Modulation of receptor recycling and degradation by the endosomal kinesin KIF16B. Cell 121, 437–450 (2005)
Soppina, V. et al. Dimerization of mammalian kinesin-3 motors results in superprocessive motion. Proc. Natl Acad. Sci. USA 111, 5562–5567 (2014)
Glotzer, M. The 3Ms of central spindle assembly: microtubules, motors and MAPs. Nature Rev. Mol. Cell Biol. 10, 9–20 (2009)
Fürthauer, M. & Gonzalez-Gaitan, M. Endocytic regulation of notch signalling during development. Traffic 10, 792–802 (2009)
Wang, S., Younger-Shepherd, S., Jan, L. Y. & Jan, Y. N. Only a subset of the binary cell fate decisions mediated by Numb/Notch signaling in Drosophila sensory organ lineage requires Suppressor of Hairless. Development 124, 4435–4446 (1997)
Fededa, J. P. & Gerlich, D. W. Molecular control of animal cell cytokinesis. Nature Cell Biol. 14, 440–447 (2012)
Korten, T. & Diez, S. Setting up roadblocks for kinesin-1: mechanism for the selective speed control of cargo carrying microtubules. Lab Chip 8, 1441–1447 (2008)
LaPointe, N. E. et al. The amino terminus of tau inhibits kinesin-dependent axonal transport: implications for filament toxicity. J. Neurosci. Res. 87, 440–451 (2009)
Stoppin-Mellet, V., Fache, V., Portran, D., Martiel, J. L. & Vantard, M. MAP65 coordinate microtubule growth during bundle formation. PLoS ONE 8, e56808 (2013)
Karpova, N., Bobinnec, Y., Fouix, S., Huitorel, P. & Debec, A. Jupiter, a new Drosophila protein associated with microtubules. Cell Motil. Cytoskeleton 63, 301–312 (2006)
Lukinavičius, G. et al. Fluorogenic probes for live-cell imaging of the cytoskeleton. Nature Methods 11, 731–733 (2014)
Goodwin, S. S. & Vale, R. D. Patronin regulates the microtubule network by protecting microtubule minus ends. Cell 143, 263–274 (2010)
Euteneuer, U. & McIntosh, J. R. Polarity of midbody and phragmoplast microtubules. J. Cell Biol. 87, 509–515 (1980)
Schiel, J. A. et al. Endocytic membrane fusion and buckling-induced microtubule severing mediate cell abscission. J. Cell Sci. 124, 1411–1424 (2011)
Januschke, J., Llamazares, S., Reina, J. & Gonzalez, C. Drosophila neuroblasts retain the daughter centrosome. Nature Commun. 2, 243 (2011)
Januschke, J. et al. Centrobin controls mother–daughter centriole asymmetry in Drosophila neuroblasts. Nature Cell Biol. 15, 241–248 (2013)
Jauffred, B. et al. Regulation of centrosome movements by numb and the collapsin response mediator protein during Drosophila sensory progenitor asymmetric division. Development 140, 2657–2668 (2013)
Ems-McClung, S. C. & Walczak, C. E. Kinesin-13s in mitosis: key players in the spatial and temporal organization of spindle microtubules. Semin. Cell Dev. Biol. 21, 276–282 (2010)
Wang, H., Brust-Mascher, I., Civelekoglu-Scholey, G. & Scholey, J. M. Patronin mediates a switch from kinesin-13-dependent poleward flux to anaphase B spindle elongation. J. Cell Biol. 203, 35–46 (2013)
Connell, J. W., Lindon, C., Luzio, J. P. & Reid, E. Spastin couples microtubule severing to membrane traffic in completion of cytokinesis and secretion. Traffic 10, 42–56 (2009)
Matsuo, M. et al. Katanin p60 contributes to microtubule instability around the midbody and facilitates cytokinesis in rat cells. PLoS ONE 8, e80392 (2013)
Yang, D. et al. Structural basis for midbody targeting of spastin by the ESCRT-III protein CHMP1B. Nature Struct. Mol. Biol. 15, 1278–1286 (2008)
Hendershott, M. C. & Vale, R. D. Regulation of microtubule minus-end dynamics by CAMSAPs and Patronin. Proc. Natl Acad. Sci. USA 111, 5860–5865 (2014)
Jiang, K. et al. Microtubule minus-end stabilization by polymerization-driven CAMSAP deposition. Dev. Cell 28, 295–309 (2014)
Gong, W. J. & Golic, K. G. Ends-out, or replacement, gene targeting in Drosophila. Proc. Natl Acad. Sci. USA 100, 2556–2561 (2003)
Gong, W. J. & Golic, K. G. Genomic deletions of the Drosophila melanogaster Hsp70 genes. Genetics 168, 1467–1476 (2004)
Seum, C. et al. Drosophila SETDB1 is required for chromosome 4 silencing. PLoS Genet. 3, e76 (2007)
Emery, G. et al. Asymmetric Rab 11 endosomes regulate delta recycling and specify cell fate in the Drosophila nervous system. Cell 122, 763–773 (2005)
Bellaïche, Y., Gho, M., Kaltschmidt, J. A., Brand, A. H. & Schweisguth, F. Frizzled regulates localization of cell-fate determinants and mitotic spindle rotation during asymmetric cell division. Nature Cell Biol. 3, 50–57 (2001)
Minestrini, G., Mathe, E. & Glover, D. M. Domains of the Pavarotti kinesin-like protein that direct its subcellular distribution: effects of mislocalisation on the tubulin and actin cytoskeleton during Drosophila oogenesis. J. Cell Sci. 115, 725–736 (2002)
Lu, B., Ackerman, L., Jan, L. Y. & Jan, Y. N. Modes of protein movement that lead to the asymmetric localization of partner of Numb during Drosophila neuroblast division. Mol. Cell 4, 883–891 (1999)
Mummery-Widmer, J. L. et al. Genome-wide analysis of Notch signalling in Drosophila by transgenic RNAi. Nature 458, 987–992 (2009)
Bitan, A., Rosenbaum, I. & Abdu, U. Stable and dynamic microtubules coordinately determine and maintain Drosophila bristle shape. Development 139, 1987–1996 (2012)
Tang, H. et al. Numb proteins specify asymmetric cell fates via an endocytosis- and proteasome-independent pathway. Mol. Cell. Biol. 25, 2899–2909 (2005)
Betschinger, J., Mechtler, K. & Knoblich, J. A. The Par complex directs asymmetric cell division by phosphorylating the cytoskeletal protein Lgl. Nature 422, 326–330 (2003)
Fabrowski, P. et al. Tubular endocytosis drives remodelling of the apical surface during epithelial morphogenesis in Drosophila. Nature Commun. 4, 2244 (2013)
Dunst, S. et al. Endogenously tagged rab proteins: a resource to study membrane trafficking in Drosophila. Dev. Cell 33, 351–365 (2015)
Bökel, C., Schwabedissen, A., Entchev, E., Renaud, O. & Gonzalez-Gaitan, M. Sara endosomes and the maintenance of Dpp signaling levels across mitosis. Science 314, 1135–1139 (2006)
Rothbauer, U. et al. A versatile nanotrap for biochemical and functional studies with fluorescent fusion proteins. Mol. Cell. Proteomics 7, 282–289 (2008)
Iwaki, T., Figuera, M., Ploplis, V. A. & Castellino, F. J. Rapid selection of Drosophila S2 cells with the puromycin resistance gene. Biotechniques 35, 482–484, 486 (2003)
Derivery, E. et al. The Arp2/3 activator WASH controls the fission of endosomes through a large multiprotein complex. Dev. Cell 17, 712–723 (2009)
Krahn, M. P., Klopfenstein, D. R., Fischer, N. & Wodarz, A. Membrane targeting of Bazooka/PAR-3 is mediated by direct binding to phosphoinositide lipids. Curr. Biol. 20, 636–642 (2010)
Goshima, G. & Vale, R. D. The roles of microtubule-based motor proteins in mitosis: comprehensive RNAi analysis in the Drosophila S2 cell line. J. Cell Biol. 162, 1003–1016 (2003)
Forer, A. & Pickett-Heaps, J. D. Cytochalasin D and latrunculin affect chromosome behaviour during meiosis in crane-fly spermatocytes. Chromosome Res. 6, 533–549 (1998)
Januschke, J. & Gonzalez, C. The interphase microtubule aster is a determinant of asymmetric division orientation in Drosophila neuroblasts. J. Cell Biol. 188, 693–706 (2010)
Loubéry, S. & Gonzalez-Gaitan, M. Monitoring notch/delta endosomal trafficking and signaling in Drosophila. Methods Enzymol. 534, 301–321 (2014)
Jauffred, B. & Bellaiche, Y. Analyzing frizzled signaling using fixed and live imaging of the asymmetric cell division of the Drosophila sensory organ precursor cell. Methods Mol. Biol. 839, 19–25 (2012)
Bell, P. B. Jr & Safiejko-Mroczka, B. Improved methods for preserving macromolecular structures and visualizing them by fluorescence and scanning electron microscopy. Scanning Microsc. 9, 843–857, discussion 858–860 (1995)
Gell, C. et al. Microtubule dynamics reconstituted in vitro and imaged by single-molecule fluorescence microscopy. Methods Cell Biol. 95, 221–245 (2010)
Lai, E. C. & Rubin, G. M. neuralized functions cell-autonomously to regulate a subset of notch-dependent processes during adult Drosophila development. Dev. Biol. 231, 217–233 (2001)
Yeh, E., Zhou, L., Rudzik, N. & Boulianne, G. L. Neuralized functions cell autonomously to regulate Drosophila sense organ development. EMBO J. 19, 4827–4837 (2000)
Gruenberg, J. The endocytic pathway: a mosaic of domains. Nature Rev. Mol. Cell Biol. 2, 721–730 (2001)
Sönnichsen, B., De Renzis, S., Nielsen, E., Rietdorf, J. & Zerial, M. Distinct membrane domains on endosomes in the recycling pathway visualized by multicolor imaging of Rab4, Rab5, and Rab11. J. Cell Biol. 149, 901–914 (2000)
Zerial, M. & McBride, H. Rab proteins as membrane organizers. Nature Rev. Mol. Cell Biol. 2, 107–117 (2001)
Bolte, S. & Cordelieres, F. P. A guided tour into subcellular colocalization analysis in light microscopy. J. Microsc. 224, 213–232 (2006)
Holtzer, L. & Schmidt, T. The tracking of individual molecules in cells and tissues (Wiley-VCH, 2010)
Tarantino, N. et al. TNF and IL-1 exhibit distinct ubiquitin requirements for inducing NEMO-IKK supramolecular structures. J. Cell Biol. 204, 231–245 (2014)
Frank, J. Three-Dimensional Electron Microscopy of Macromolecular Assemblies: Visualization of Biological Molecules in Their Native State 2nd edn (Oxford University Press, 2006)
Meijering, E., Dzyubachyk, O. & Smal, I. Methods for cell and particle tracking. Methods Enzymol. 504, 183–200 (2012)
Kural, C. & Kirchhausen, T. Live-cell imaging of clathrin coats. Methods Enzymol. 505, 59–80 (2012)
Thorn, K. S., Ubersax, J. A. & Vale, R. D. Engineering the processive run length of the kinesin motor. J. Cell Biol. 151, 1093–1100 (2000)
Mayer, B., Emery, G., Berdnik, D., Wirtz-Peitz, F. & Knoblich, J. A. Quantitative analysis of protein dynamics during asymmetric cell division. Curr. Biol. 15, 1847–1854 (2005)
Kirchhofer, A. et al. Modulation of protein properties in living cells using nanobodies. Nature Struct. Mol. Biol. 17, 133–138 (2010)
Derivery, E., Helfer, E., Henriot, V. & Gautreau, A. Actin polymerization controls the organization of WASH domains at the surface of endosomes. PLoS ONE 7, e39774 (2012)
Vonderheit, A. & Helenius, A. Rab7 associates with early endosomes to mediate sorting and transport of Semliki forest virus to late endosomes. PLoS Biol. 3, e233 (2005)
We thank A. Gautreau, in whose laboratory the GBP was cloned, for allowing us to use it before his own publication. We thank A. Gautreau, R. Vale, A. Houdusse, A. Roux, A. Trushko and V. Stoppin-Mellet for discussions and advices. We thank N. Chiaruttini for the preparation of liposomes and for insights throughout the project. We thank R. Le Borgne, S. Eaton, S. De Renzis, F. Karch, W. Zhong and R. Stanewsky for flies. We thank P. Kapusta from Picoquant for help with FLIM data analysis. We thank V. Stoppin-Mellet, M. Vantard and J. Gaillard for the gift of GFP-MAP65-1. We are indebted to J. Teyssier, E. Allémann and N. Boulens for letting us use their scanning electron microscopes. E.D. is supported by a Long term Fellowship of the Human Frontier Science Program. L.H. and S.L. are supported by a Marie-Curie Intra-European Fellowship. M.G.G. is supported by Departement de l’Instruction Publique of the Canton of Geneva, the SNF, the SystemsX epiPhysX program, the NCCR Chemical Biology program, the ERC (Sara and Morphogen) and the Polish–Swiss research programs.
The authors declare no competing financial interests.
Extended data figures and tables
a, b, Characterization of the polyclonal anti-Klp98A antibody in western blot and immunofluorescence. a, Western blot of total extracts of w1118 and Klp98A− mutant L2 larvae using anti-Klp98A antibodies. Klp98A is the major band above 180 kDa that disappears in Klp98A− mutants. Western blot is representative of two experiments. b, Confocal image showing SOPs in the notum labelled by iDelta20 and Klp98A immunofluorescence in w1118 and Klp98A− mutants. Imaging was performed by spinning-disk confocal microscopy (SDCM, z projection). Dashed line, SOP outline. The Klp98A staining associated with Sara endosomes disappears in Klp98A− mutants. c–e, Klp98A-containing vesicles are early endosomes. c, SOPs in metaphase showing Klp98A (first row; either endogenous, in the two first columns, or, in the other columns, Klp98A–mCherry at low expression levels, using temperature control) together with endosomal markers (second row; iDelta20, GFP–Sara at low expression levels, as well as GFP–Rab5 and YFP–Rab7 and YFP–Rab11 at endogenous expression levels). Third row, merge; red channel, Klp98A; green channel, endosomal marker; mRFP–Pon, in blue, is also shown in some panels. SDCM imaging. Except for YFP–Rab11, fixed samples were imaged (z projection of the entire cell; maximum intensity). Insets correspond to single planes of higher magnification images of a representative endosome (arrow). For YFP–Rab11, images correspond to one single z plane in a live cell. Dashed lines, cell outline. Note an apparent lack of fine-grain co-localization between Klp98A and iDelta20, which reflects the existence of distinct sub-domains within a particular vesicular compartment, as previously reported for early endosomes57,68,69. Scale bars in b and c are 5 μm. d, Automated quantification of the 3D co-localization (see co-localization section in Methods) of Klp98A with the various endosomal markers shown in c. Quantification is performed in 2D for YFP–Rab11. Klp98A shows a high level of co-localization with GFP–Sara and GFP–Rab5. A lower, but significant, level of co-localization is observed with YFP–Rab7, and almost no co-localization with YFP–Rab11 (Kruskal–Wallis non-parametric ANOVA followed by Dunn’s post-hoc test; N, number of cells; n, number of endosomes). e, Quantification of the number of iDelta20 endosomes per SOP in control cells and cells expressing Klp98A–mCherry at low levels. Klp98A–mCherry expression does not affect the number of iDelta20 endosomes per cell (Mann–Whitney rank sum test; N, number of cells). f, Western blot anti-Protein-C tag (PC) of the input and of floating fractions from liposome floating assays using small unilamellar vesicles (SUVs) of different compositions (DOPC:DOPS 90:10; DOPC:DOPS:PI(3)P 80:10:10; DOPC:DOPS:PI(4)P 80:10:10; DOPC:DOPS:PI(5)P 80:10:10) and purified Klp98A–GFP–PC (see Methods). Floating fractions corresponding to equal amounts of liposomes were loaded. Note that, to show the input together with the other samples in this image, the input lane is displayed after a lower exposure time and separated from the other lanes by a vertical line. Klp98A–GFP–PC directly binds to PI(3)P-containing liposomes, and does not bind to PI(4)P- nor PI(5)P-containing liposomes. Western blot is representative of two experiments. g–i, Generation of the Klp98AΔ47 deletion mutant by homologous recombination (see Methods). g, Klp98A isoform A (Klp98A-RA) gene organization. Homology regions used to generate the mutant as well as PCR primers for its characterization are displayed. h, Gene organization after homologous recombination: the Klp98A gene has been replaced by an AttP site. i, PCR analysis of control (w1118) and the Klp98AΔ47 mutant showing the expected short amplicon in the Klp98AΔ47 mutant (5′/3′ primers) and the presence of the Attp site in the Klp98AΔ47 mutant locus (Attp/3′ primers). j, Sequence of the three Klp98A alleles obtained by zinc-finger-mediated mutagenesis (see Methods). Position refers to the coding sequence of Klp98A (CG5658-PA). Klp98AΔ6 corresponds to a missense mutation followed by a six-base-pair deletion causing missense change and deletion of two amino acids in the protein. Klp98AΔ7 and Klp98AΔ8 correspond to seven- and eight-base-pair deletions causing frame shifts. All the Klp98A mutants generated in this study are viable and fertile in trans with Klp98AΔ47. k, Western blot of total extracts of w1118 control and Klp98A− mutant L2 larvae using anti-Klp98A antibodies. n.s., non-specific bands (see a) serving as loading controls. Full length Klp98A is absent in both Klp98AΔ7 and Klp98AΔ8 mutants, but is produced at normal levels in Klp98AΔ6. Western blot is representative of two experiments. For source gel images of the various western blots displayed in this figure, see Supplementary Fig. 1.
Extended Data Figure 2 iDelta20 co-localizes with Sara and they both show the same dynamics in all the genotypes considered in this study.
a, b, iDelta20 co-localizes with Sara in all the genotypes considered in this study. a, SOPs in metaphase showing GFP–Sara or RFP–Sara (first row) and iDelta20 (second row) in the different backgrounds studied in this work (SDCM imaging; image corresponds to maximum-intensity z projection of entire cells; insets corresponds to single planes). Third row, merge: Sara in red; iDelta20 in green; mRFP–Pon, mCherry–Pon, Klp98A–mCherry or GFP–Patronin in blue. Except for the right-most panel (GBP experiment), GFP–Sara was expressed at low levels using an ase > GFP–Sara construct, yielding Sara endosomes which appear mostly as diffraction-limited vesicles. Dashed line, cell outline (pIIb blue, pIIa white). For the right-most panel, RFP–Sara expression is at some higher level under the control of Neur-Gal4. b, Automated quantification of the 3D co-localization of iDelta20 and fluorescent-protein-tagged Sara in the different backgrounds (see Methods). The degree of colocalization of the iDelta20 signal in Sara–FP vesicles was, as in control, around 80% amongst all the genotypes studied here (Kruskal–Wallis non parametric ANOVA followed by Dunn’s post-hoc test; N, number of cells; n, number of endosomes). Indeed, for all the genotypes, differences in the percentage of co-localization with respect to the control were not statistically significant, with exception of PatroninRNAi (asterisk, P < 0.05) where the co-localization is slightly lower than in control (about 12% difference). c–h, Dynamics of FP–Sara endosomes during SOP mitosis in all the genotypes considered in this study. Dividing SOPs showing FP–Sara endosomes in the different backgrounds studied in this report (SDCM imaging; maximum-intensity z projection, except for GFP–Patronin channel in h, which corresponds to single planes). Imaging was performed in the animal, without dissection. The dynamics of FP–Sara endosomes (spindle targeting or not, asymmetric segregation or not, and so on) parallels what we report for iDelta20 endosomes throughout this study. i, Dynamics of the percentage of Sara endosomes at the central spindle and of endosome asymmetry measured simultaneously in control SOPs (dark line, mean; lighter area, s.e.m.). Sara endosomes become asymmetric upon spindle targeting. Time was registered between videos (see Methods) and time point zero corresponds to anaphase B onset. n, number of cells. Note that, for comparison, the plot of percentage of Sara endosomes at the central spindle shown here is the same as Fig. 1f. Scale bars, 5 μm.
Extended Data Figure 3 Klp98A mutants affects asymmetric cell fate assignation in conditions where Neuralized or Numb are compromised.
See also ‘Rationale and quantification of the Neur and Numb phenotypes’ in Methods. a–c, Klp98A mutants show asymmetric cell fate assignation phenotypes in conditions where Neuralized is compromised. a, Scanning electron microscopy (SEM) images of adult w1118 control, pnr > NeurRNAi Klp98AΔ47/Klp98AΔ6 double mutant flies and pnr > NeurRNAi Klp98AΔ47/+ sibling controls (see ‘Detailed genotypes’ in Methods). Yellow dashed lines delimit the panier expression region corresponding to the region between the right and left pairs of dorso-central macrochaetes. For each genotype, the image corresponds to the most frequent phenotypic class in the population (asterisk in the statistical distribution shown in b). Scale bar, 200 μm. b, Number of organs in the panier expression region scored on SEM images for the genotypes shown in a (for scoring criteria, see ‘Quantification of the Neur and Numb phenotypes’ in Methods). Five phenotypic classes of equal range of bristles (pc1–pc5) were considered, spanning from 0 bristles (stronger phenotype) to 130 bristles (maximum number of bristles scored in control flies). n, number of flies scored. The number of lineages which generated bristles is significantly reduced in pnr > NeurRNAi Klp98AΔ47/Klp98AΔ6 double mutant flies compared to Klp98AΔ47/+ heterozygous sibling controls (χ2 test, P < 0.05). c, SOPs are specified in pnr > neurRNAi Klp98AΔ47/Klp98AΔ8 double mutants. Neuron-specific elav staining of nota in pnr > neurRNAi Klp98AΔ47/+ sibling controls (inside and outside the panier expression region) and of pnr > neurRNAi Klp98AΔ47/Klp98AΔ8 double mutants (inside the panier expression region). Note that neurons are present in pnr > neurRNAi Klp98AΔ47/Klp98AΔ8 double mutants, indicating that in this mutant SOPs were specified. Scale bar, 10 μm. d–f, Klp98A mutants suppress cell fate assignation phenotypes in hypomorphic Numb mutants. d, SEM images of adult w1118 control, Numb2/NumbSW; Klp98AΔ8/Klp98AΔ47 double mutant and its Numb2/NumbSW; Jupiter-GFP/Klp98AΔ47 sibling control, pnr > numbRNAi Klp98AΔ47/+ control and pnr > numbRNAi Klp98AΔ47/Klp98AΔ8 double mutant. In this panel and as well as e and f, we used a Jupiter–GFP knock-in as a control chromosome. Arrows indicate organs with multiple sockets, characteristic of Numb mutants. This phenotype is suppressed in Klp98A mutants. Scale bar, 100 μm. Yellow dashed lines delimit the panier expression region where phenotypes were scored (also in the NumbSW conditions). e, Percentage of total organs showing multiple sockets in the panier expression region, scored on SEM images in Numb2/NumbSW; Klp98AΔ6/Klp98AΔ47 double mutants (and its Numb2/NumbSW; Jupiter-GFP/Klp98AΔ47 sibling controls), Numb2/NumbSW; Klp98AΔ8/Klp98AΔ47 double mutants (and its Numb2/NumbSW; Jupiter-GFP/Klp98AΔ47 sibling controls) and Numb15/NumbSW; Klp98AΔ6/Klp98AΔ47 double mutants (and its Numb15/NumbSW; Klp98AΔ47/TM6B or Numb15/NumbSW; Klp98AΔ6/TM6B sibling controls). All double mutants show a significant reduction of the percentage of multiple sockets when compared to their respective sibling control (χ2 test, P < 0.001). n, total number of microchaete scored. f, Percentage of total organs showing multiple sockets in the panier expression region in pnr > numbRNAi Klp98AΔ47/+ controls, pnr > numbRNAi Klp98AΔ47/Klp98AΔ6 double mutants and pnr > numbRNAi Klp98AΔ47/Klp98AΔ8 double mutants. Both mutants show a significant reduction of the percentage of multiple sockets when compared to the control (χ2 test, P < 0.001).
a–e, Determination of the endosome diffusion coefficient (see also ‘Mean square displacement analysis of endosome tracks’ in Methods). a–b, MSD individual traces as a function of delay time for each endosome track: n = 103 tracks in control SOPs (a) and n = 158 in Klp98AΔ47/Δ8 SOPs (b). The thick black line corresponds to the weighted mean curve61, which weighs the MSD curves according to their certainty (see Methods). c, Parabolic or linear fitting of the weighted MSD as a function of delay time in control and Klp98A− SOPs (for details see Methods). Dark line, weighted mean; lighter area, s.e.m. In the absence of motor activity in the Klp98A null mutant, a linear fit reflects diffusive movement. In the control, a parabolic fit reflects a directed movement as a consequence of Klp98A motor activity. d, e, Histogram of instantaneous velocities of Sara endosomes in Klp98AΔ47/Δ8 mutant SOPs along x (d) and y (e) axes. f–i, Precision of the Pavarotti tracking algorithm. f, Scheme showing the two reference frames (image reference frame and central spindle reference frame) considered in the tracking analysis. For details see ‘Spatio-temporal registration’ in Methods. g–i, Histograms of the deviation from the mean value of α (g), xc (h) and yc (i) from videos of fixed tissue. FWHM values shown in the histograms correspond to the accuracy of each parameter. j–n, Accuracy of our iDelta20 detection and tracking method. j–l, 100 iDelta20 endosomes were tracked in fixed material and the deviation to their mean x and y position was computed. Deviations are represented either as a 2D density plot (j) or decomposed in histograms along the x axis (k) and the y axis (l). FWHM measurements indicate a positional accuracy of 57 nm on the x axis and 53 nm on the y axis, respectively. m, Dependence of the tracking accuracy on the endosome signal-to-noise ratio (SNR) using videos of fixed tissue (see Methods). To ensure reliable iDelta20 tracking, we only considered endosomes with an SNR > 15 (dashed line). n, SNR of a single iDelta20 endosome from a video of a fixed tissue imaged under constant laser illumination. Each time point corresponds to a 3D stack of seven slices through the central spindle. The SNR remains constant throughout iterative laser excitation of Atto647N fluorophore, indicating high photostability of this fluorophore and, therefore, that iDelta20 tracking accuracy does not decrease over time due to bleaching. o, p, Principle of the time registration method using the PW dynamics. Since the temporal profile of Pavarotti contraction is stereotypic, each cell can be aligned in time with a reference cell by determining the time delay τcell that needs to be applied to this cell of interest in order to minimize the difference, in absolute value, between the two temporal profiles (o) (that is, find the τcell for which is minimum (p)). q, t, Unregistered PW (q) and PL (t) curves of 45 cells. r, u, Same data set as in q and t, but upon time registration. In this panel and others in this report, the time point zero in registered time corresponds to anaphase B onset (that is, when PW starts to contract). s, v, Same plot as in r and u, but represented as a density plot to show the collapse of the temporal profiles upon registration (rainbow lookup table). Note that the registered PL curves (which were not used in the registration process based on PW) also collapse, validating the method. w, Average temporal contraction profile of PW and ‘Jupiter gap’ (see Methods). Since Jupiter is excluded from the region that Pavarotti occupies (see Fig. 3a), the absence of Jupiter (Jupiter gap, defined as a FWHM) can be used as a proxy of the Pavarotti region. The PW and Jupiter gap temporal profiles are very similar, thus either marker can be used for temporal registration. n, number of cells. x, Average PW/Jupiter gap temporal profiles in control, Patronin RNAi#1, Klp10A RNAi and Klp98A mutant conditions. The profiles are very similar: time registration can be applied to Patronin and Klp10A RNAi conditions, as well as to Klp98A mutants. n, number of cells. y, Examples of bidirectional movements on the central spindle. iDelta20 endosomes were automatically tracked during mitosis of GFP–Pavarotti expressing cells. Movement of nine endosomes on the x axis of the central spindle is shown. Green dashed line indicates PL over time. iDelta20 tracks are mostly confined within the Pavarotti-positive region containing the antiparallel overlap until their departure. Tracks explore both sides of the Pavarotti positive region, but do it more often and deeper on the pIIa side.
a, b, Generation of the ‘average SOP cell’. a, Principle of the spindle averaging technique. Dividing SOPs expressing GFP–Pavarotti and Jupiter–mCherry were imaged by SDCM (z projection). Pavarotti signal was subsequently tracked to register all the movies in space and time before averaging. b, Time frames from a time-lapse average video of nine dividing SOPs. The central spindle is asymmetrical, with an enrichment of Jupiter density on the pIIb side (left). Elapsed time in seconds. c–i, The central spindle of SOPs is asymmetric. c, Left two panels: illustration of the two methods (pseudo-line-scan and segmentation method) used to measure central spindle asymmetry (see ‘Measurement of spindle asymmetry in anaphase’ in Methods). Right panel, Jupiter–mCherry enrichment in pIIb as scored with either method. Samples are the same as in Fig. 3a. Both methods gave comparable results when applied to the samples in Fig. 3a, but also to immunofluorescence of endogenous tubulin in fixed samples (see h, i). d–g, Average central spindle images (d, f) and line scan (e, g; line scan is based on the ROI shown in the top panels in d of both SOP and neighbouring epithelial cells (non-SOP) immunolabelled using Oregon Green-514 anti-β-tubulin antibodies and Atto647N anti-acetylated-α-tubulin antibodies. SDCM imaging with sum projection to generate an average cell (see ‘Image averaging’ chapter in Methods). Anterior to the left as determined by the mRFP–Pon signal (not shown). Central spindles of SOPs are asymmetric, with an enrichment of tubulin and a corresponding enrichment of acetylated-tubulin on the pIIb side of the spindle, while non-SOP central spindles are symmetric. h, i, Tubulin and acetylated tubulin enrichment in the anterior pIIb cell in SOPs and anterior cells in non-SOP mitoses in the samples presented in d and f using either the pseudo-line-scan or the segmentation methods as in c. Note that the non-SOP cells do not show a statistically significant enrichments (zero enrichment is within the s.e.m. of the samples), while SOP cells do. Confirming this, a Kruskal–Wallis non-parametric ANOVA followed by Dunn’s post hoc test (P < 0.05) shows that in SOP cells the anterior (pIIb) enrichment of microtubule markers is higher than in non-SOP cells. These statements are true whether the line scan or the segmentation methods are considered. n = 47 for non-SOP and 57 for SOP cells. j, k, Spindle asymmetry occurs downstream of Par complex signalling. j, Representative dividing control or lgl3A-expressing SOPs showing mRFP–Pon and Jupiter–GFP. Upon lgl3A overexpression, Pon segregation becomes symmetrical. Central spindle asymmetry (Jupiter–GFP channel) is also abolished upon lgl3A expression. Note that the control panel is the same as Fig. 3g, shown here for comparison. k, Jupiter–GFP enrichment in anterior pIIb daughters (control) or in anterior daughters (lgl3A) at late cytokinesis. P < 0.001 (Mann–Whitney test). l–s, Analysis of spindle asymmetries from early- to late-mitotic phases (see also ‘Centrosome asymmetry’ in the Supplementary Discussion). l–n, Spindle asymmetry occurs late during SOP mitosis. Dividing SOPs showing Jupiter–GFP, GFP–Patronin or SiR-tubulin in metaphase (l) or late anaphase (m, ∼600 s registered time). Antero-posterior orientation is based on the mRFP–Pon cortical signal (not shown). n, Jupiter–GFP, GFP–Patronin or SiR-tubulin enrichment in the pIIb side of the spindle (; ρa, pIIa density; ρb, pIIb) in metaphase or late anaphase. While the spindle is not asymmetric in metaphase, in late anaphase, the density of Jupiter–GFP, GFP–Patronin and SiR-tubulin is enriched in the pIIb side compared to pIIa. Statistics were performed using a Kruskal–Wallis non-parametric ANOVA followed by Dunn’s post hoc test (P < 0.05). o, p, Centrosome asymmetry in anaphase is unaffected upon Patronin or Klp10A depletion. o, SOPs in late anaphase showing Jupiter–GFP in control, Patronin-depleted or Klp10A-depleted conditions. The centrosome in pIIa (white arrow) is brighter than that in pIIb (orange arrow) in all conditions. p, Intensity ratio of Jupiter–GFP signal in the pIIa centrosome over the pIIb centrosome in control, Patronin- and Klp10A-depleted SOPs in late anaphase (mean ± s.e.m.). In control, the enrichment of Jupiter–GFP in the pIIa centrosome is statistically significant (the s.e.m. does not cross the ratio = 1 line). In Patronin and Klp10A depletion this asymmetry is not statistically different to that of control (Kruskal–Wallis non-parametric ANOVA followed by Dunn’s post hoc test). Note that in control cells, in the central spindle, it is the anterior pIIb (not the posterior pIIa) side that shows a higher density of microtubules. This observation untangles the asymmetry of the centrosomes from the asymmetry of the central spindle. Consistent with this, in Patronin depletion and Klp10A depletion conditions, central spindle asymmetry is abolished (see here, but also Fig. 3e, g, h), while centrosome asymmetry is not affected. q–s, Centrosome asymmetry can be detected before anaphase. Dividing SOPs showing Jupiter–GFP, GFP–Patronin, SiR-tubulin and EB1–GFP in prophase (q) or prometaphase (r). All four markers show a signal enrichment in one centrosome compared to the other, suggesting different microtubule-organizing activities of the two centrosomes. s, Intensity ratio of the centrosome-associated signals of Jupiter–GFP, GFP–Patronin, SiR-tubulin and EB1–GFP in prophase and prometaphase. Here, the pIIa/pIIb identity of the centrosome could not be determined since the mitotic spindle rotates in metaphase. We thus plotted the ratio of the brightest over the dimer centrosome. This suggests that centrosome asymmetry is acquired early, perhaps due to centrosome age. However, it must be noted that centrosome asymmetry is untangled from central spindle asymmetry. Individual data points are figured when number of cell scored is inferior to 5. All images in this figure correspond to SDCM imaging (maximum-intensity projection). Scale bars, 5 μm. Lower panels in b–d, f, j–m, o, q, r correspond to rainbow lookup table. n, number of cells scored.
a, Validation of the anti-Patronin antibody and of the depletion of Patronin by the Patronin RNAi#1 sequence. Anti-Patronin western blot of total extracts of S2 cells treated with GFP RNAi or Patronin RNAi#1. Patronin corresponds to the major band above 180 kDa that disappears upon RNAi treatment, validating both the RNAi sequence and the antibody. The Patronin RNAi construct presented in the main text corresponds to this Patronin RNAi #1. Asterisks indicate non-specific bands. Western blot is representative of two experiments. For source gel images, see Supplementary Fig. 1. b, c, Validation, in S2 cells, of the Patronin RNAi#1 sequence used in the SOP. b, Upper panel, α-tubulin immunofluorescence of S2 cells treated with GFP RNAi or Patronin RNAi#1 in metaphase. Lower panel, Jupiter–GFP signal control and Patronin RNAi#1 depleted SOPs in metaphase. Live imaging. c, Measurement of the spindle pole to spindle pole distance of metaphase spindles in the samples described in b. Statistics were performed using a Kruskal–Wallis non-parametric ANOVA followed by Dunn’s post hoc test; P < 0.05. Similar to what was reported by Goodwin and Vale (ref. 16) using a different RNAi sequence, depletion of Patronin in S2 cells by our Patronin RNAi#1 leads to short and disorganized mitotic spindles. This is however not the case in the SOP: the mitotic spindle is neither short nor disorganized. The difference between the SOP and S2 cells can be explained by the different levels of expression of the RNAi in the two systems: in the SOP, a milder Patronin depletion does allow the cell to proceed to anaphase without metaphase spindle defects/arrest. Later, the SOP central spindle seems to be more sensitive than the metaphase spindle and it is affected by our conditions of Patronin depletion in the SOP: it becomes symmetric (see below and Fig. 3e, g, h). d, e, Average central spindle images (d) and line scan (e; line scan is based on a ROI, as shown in Extended Data Fig. 5c) of both control and Patronin-depleted SOPs immunolabelled using Atto647N anti-acetylated-α-tubulin antibodies. SDCM imaging with sum projection to generate an average cell (see ‘Image averaging’ in Methods). Anterior to the left as determined by the mRFP–Pon signal (not shown). Spindle asymmetry is abolished upon Patronin depletion by two independent RNAi sequences. Patronin RNAi#1 is referred as Patronin RNAi in the main text. Note that the control line scan is the same as Extended Data Fig. 5g, shown here for comparison. f, Acetylated tubulin enrichment in the pIIb cell in control and Patronin-depleted SOPs measured by the pseudo-line-scan method (see Extended Data Fig. 5c). pIIb microtubule enrichment is defined as , with Ib and Ib the respective pIIb and pIIa peak values of the linescans shown in e. The asymmetry of the central spindle is significantly reduced upon Patronin depletion (Kruskal–Wallis non-parametric ANOVA followed by Dunn’s post hoc test; P < 0.05). Note that the control data set is the same as Extended Data Fig. 5h, shown here for comparison. g, h, Specificity of the central spindle symmetry phenotype by Patronin RNAi#1 as shown in Fig. 3e, g, h: rescue of the effects of RNAi-induced Patronin depletion by overexpression of GFP–Patronin. g, Representative dividing SOPs expressing mRFP–Pon and GFP–Patronin (GFP–Patronin control; left panel) or, in addition, expressing Patronin RNAi#1 (GFP–Patronin + Patronin RNAi#1; middle and right panels) at late cytokinesis. SDCM imaging, z projection. RNAi-mediated depletion is rescued by GFP–Patronin overexpression mediated by the UAS/Gal4 system. Note that, in the RNAi conditions, the GFP–Patronin signal is reduced, reflecting the fact that the GFP–Patronin mRNA is targeted by RNAi; see, for example, GFP–Patronin signal in central versus left panels. Spindle asymmetry is rescued upon overexpression of GFP–Patronin in the RNAi condition (see right panel, where the brightness has been enhanced to visualize the different microtubule levels of the spindle). h, GFP–Patronin enrichment in the pIIb cell in control and rescued SOPs as in g. Enrichment in both conditions is not statistically significant (P = 0.175; Mann–Whitney test). i, j, Co-depletion of Klp10A and Patronin suppresses their respective phenotype on central spindle asymmetry. i, Central spindle in dividing control, Patronin-, Klp10A-depleted and Patronin/Klp10A-co-depleted SOPs expressing Jupiter–GFP (SDCM, z projection). Antero-posterior orientation is based on the mRFP–Pon signal (not shown). Note that, upon co-depletion of Patronin and Klp10A, the phenotype of symmetric central spindle observed in Patronin and Klp10A RNAi conditions is suppressed to recover the levels of asymmetry observed in the control situation (see lookup table image; bottom panels). j, Quantification of GFP–Jupiter signal enrichment in the pIIb cell in control and depleted/co-depleted SOPs as in i. The asymmetry of the central spindle is significantly reduced upon Patronin or Klp10A depletion, but restored to control levels upon co-depletion of Patronin and Klp10A (Kruskal–Wallis non-parametric ANOVA followed by Dunn’s post hoc test; P < 0.05). Note that the control data set is the same as Extended Data Fig. 5k, shown here for comparison. Lower panels in d, g and i correspond to the rainbow lookup table. n, number of cells scored. Scale bars, 5 μm (b, g) and 2 μm (i).
Extended Data Figure 7 Impaired spindle asymmetry and asymmetric Sara endosome segregation upon Patronin depletion.
a–e, Endosome motility analysis in Patronin RNAi#1 condition. a, Spatio-temporal density plot of iDelta20 endosomes at the central spindle in Patronin-depleted cells. iDelta20 endosomes were automatically tracked during mitosis of ten cells expressing GFP–Pavarotti and Patronin RNAi#1. After spatial registration of each track with respect to its own central spindle reference frame and temporal registration using PW constriction, the resulting 1,541 tracks were all plotted as a spatio-temporal density map (x axis, time). Overlaid continuous white line, averaged Pavarotti length (PL); dashed line, centroid of the Pavarotti region. iDelta20 tracks are enriched within the Pavarotti-positive region. In contrast to control, tracks explore approximately equally both sides beyond the Pavarotti-positive region. b, c, Representative endosome track, decomposed on the x (b) and y (c) axes. On the x axis, movement is bidirectional and mostly confined within the Pavarotti-positive region. On the y axis, movements are limited or at most follow the contraction of the PW. d, Velocity of microtubule-based motility towards the pIIa or the pIIb cells during bidirectional movement on the x axis. Speed does not depend on orientation of the movement (n = 149 strides in 45 tracks; non-significant in Mann–Whitney test). e, Residence time of endosomes within both cells during bidirectional movement. In contrast to the control, there is no bias (n = 30 endosome tracks; non-significant in Mann–Whitney test). f, Dynamics of the percentage of iDelta20 endosomes at the central spindle in control, Patronin- and Klp10A-depleted SOPs as in Fig. 1f (dark line, mean; lighter area, s.e.m.). iDelta20 endosomes reach the central spindle in Patronin- and Klp10A-depleted cells with similar kinetics than in control cells. Note that the control data set is the same as Fig. 1f, shown here for comparison. g, h, Specificity of the endosomal targeting symmetry phenotype induced by Patronin RNAi#1 as shown in Fig. 4a, b: overexpression of GFP–Patronin rescues the effects of Patronin RNAi#1 on Sara endosome asymmetric segregation. g, Representative dividing SOP showing mRFP–Pon and iDelta20 (control), or, in addition, expressing Patronin RNAi#1, or Patronin RNAi#1 and GFP–Patronin (GFP–Patronin + Patronin RNAi#1, third and fourth panels) at late cytokinesis. SDCM imaging, z projection. Sara endosome symmetric segregation induced by Patronin RNAi#1 expression is rescued by overexpression of GFP–Patronin. Note that spindle asymmetry is also rescued (fourth panel), as shown above in Extended Data Fig. 6g, h. h, Percentage of iDelta20 in the pIIa daughter cell at late cytokinesis in control, Patronin RNAi#1-treated and rescued cells (GFP–Patronin + Patronin RNAi#1). GFP–Patronin expression rescued the symmetric iDelta20 segregation induced by the Patronin RNAi#1 (ANOVA test followed by Holm–Sidak post hoc test). i, j, Confirmation of the effects of Patronin depletion on Sara endosome asymmetric segregation by an independent RNAi sequence, Patronin RNAi#2. i, Representative control or Patronin RNAi#2-treated SOPs showing mRFP–Pon and iDelta20 at late cytokinesis. SDCM imaging, z projection. j, Percentage of iDelta20 in pIIa at late cytokinesis in control and Patronin RNAi#2-treated cells. iDelta20 percentage in pIIa is significantly lower upon Patronin depletion (P < 0.05; unpaired two-tailed t-test). k, l, Co-depletion of Klp10A and Patronin suppresses their respective phenotype on endosome asymmetry. k, Control, Patronin-, Klp10A-depleted and Patronin/Klp10A-co-depleted SOPs showing mRFP–Pon and iDelta20 in late cytokinesis. SDCM imaging, z projection. l, Percentage of iDelta20 in the pIIa daughter cell at late cytokinesis in control, Patronin-, Klp10A-depleted and Patronin/Klp10A-co-depleted SOPs. Note that the control, Patronin RNAi and Klp10A RNAi data sets are the same as Fig. 4b, shown here for comparison. Endosome segregation becomes symmetrical upon Patronin or Klp10A depletion, but is restored to control levels upon co-depletion of Patronin and Klp10A (P < 0.05; ANOVA followed by Tukey’s post hoc test). Note that spindle asymmetry is also suppressed (Extended Data Fig. 6i, j). Unless stated otherwise, n corresponds to the number of SOP scored. Scale bars, 5 μm.
a, Model of spindle asymmetry generation by the Patronin/Klp10A machinery. Directly or indirectly, Par complex signalling leads to a biased localization of Patronin, which is thereby enriched on the pIIb side of the central spindle (left panel). Patronin antagonizes Klp10A microtubule depolymerization activity and therefore net microtubule depolymerization activity is stronger on the pIIa side of the spindle (middle panel). As a consequence, the central spindle becomes asymmetric with a higher density in the pIIb side (right panel). b, Cartoon of the rationale of the ‘nanobody assay’ for central spindle inversion (see also ‘Rationale of the nanobody experiment’ in the Supplementary Discussion). Expression of an anti-GFP nanobody (GBP) fused to an anterior cortex localization domain (Pon; left panel) leads to the specific targeting of GFP-tagged molecules at the anterior cortex. Due to the high affinity of the GBP for GFP66, most of the GFP–Patronin molecules are bound to the nanobody (see Extended Data Fig. 9a–d). Owing to the high off-rates of Pon for the cortex67 and Patronin for the spindle (see Extended Data Fig. 9e) in metaphase, the Pon–GBP/GFP–Patronin complex is in three-state equilibrium between the cytosol, the anterior cortex and the spindle (left panel). In late anaphase, owing to the diffusion barrier at the central spindle (because of crowding), the situation becomes different in pIIa and pIIb: while in pIIb there is a three-state equilibrium (anterior cortex, cytosol, spindle), in pIIa there is only a two-state equilibrium (cytosol, spindle; middle panel). This generates an asymmetry of the amount of Patronin at the spindle: in pIIb there is an additional sink provided by the anterior cortex. As a consequence, the antagonistic activity of Patronin on Klp10a-mediated microtubule depolymerization is depleted in pIIb, thereby biasing microtubule density to higher levels in pIIa (right panel). c, Dividing SOPs expressing mRFP–Pon and GFP–Patronin (GFP–Patronin control) or, in addition, expressing GBP–Pon (GFP–Patronin + GBP–Pon) at late cytokinesis. Like in control cells (upper panel), polarized mRFP–Pon localization is observed upon co-expression of GBP–Pon and GFP–Patronin (middle panel) but, occasionally, low levels of mRFP–Pon are also detected in the pIIa cell cortex (lower panel). Note that the GFP–Patronin control panel is the same as Extended Data Fig. 6g, shown here for comparison. d, Frequency of the leaking pIIa mRFP–Pon signal upon expression of the indicated fusion proteins. Cells showing leaking Pon cortical localization in pIIa were excluded from the subsequent analysis of spindle and endosome asymmetries considered in this report. e–g, Analysis of spindle inversion in the nanobody assay using a battery of microtubule markers. e, Enrichment in pIIb of GFP–Patronin in GFP–Patronin control and GFP–Patronin + GBP–Pon cells. Co-expression of GFP–Patronin and GBP–Pon induces a shift of the frequency distribution of GFP–Patronin asymmetry towards enrichment in pIIa and lower levels of enrichment in pIIb. In these conditions, inverted spindles are observed, which are only very rarely seen in control cells. f, Dividing control or GFP–Patronin and GBP–Pon expressing SOPs stained by anti-acetylated-α-tubulin (Ac-tubulin) immunofluorescence. g, Enrichment in pIIb of Ac-tubulin in control and GFP–Patronin + GBP–Pon cells. Note the shift towards pIIa enrichment induced by coexpression of GFP–Patronin and GBP–Pon. h, Percentage of iDelta20 in the pIIa daughter cell at late cytokinesis upon expression of the indicated fusion proteins. Co-expression of GBP–Pon with GFP–Patronin causes a statistically significant reduction of iDelta20 asymmetry compared to control. *P < 0.001 compared to control, as calculated by ANOVA followed by Tukey’s post hoc test. Comparisons between the other genetic backgrounds and the control or between themselves are not significant. All images in this figure correspond to SDCM imaging (maximum-intensity projection). n, number of SOPs scored. Scale bars, 5 μm.
a–d, Most GFP–Patronin in the SOP is bound to GBP–Pon. a, Dividing SOPs expressing mRFP–Pon and GFP–Patronin (left panels) or GBP–mCherry–Pon and GFP–Patronin (right panels) were imaged by SDCM (maximum-intensity projection). Note that GBP–mCherry–Pon colocalizes with GFP–Patronin both at the cortex and at the spindle. b, GBP binding increases the fluorescence lifetime of GFP in vitro. Purified (His)6–GFP was coupled to TALON beads through the (His)6 tag (left panels) or to GBP beads through GFP itself (right panels). The GFP fluorescence lifetime was imaged by FLIM (see Methods). Bottom panels correspond to images where abundance of GFP (pixel intensity) is combined with (multiplied by) the lifetime of GFP (pixel colour). Note that lifetime increases (that is, more red pixels) in the GBP beads when compared to the TALON beads (more cyan pixels). c, GBP binding increases the fluorescence lifetime of GFP–Patronin in vivo. Dividing SOPs expressing mRFP–Pon and GFP–Patronin (GFP–Patronin, control) or expressing GBP–Pon in addition (GFP–Patronin + GBP–Pon) at metaphase imaged by FLIM. As in b, bottom panel corresponds to images where abundance of GFP–Patronin (pixel intensity) is combined with its lifetime (pixel colour). Note that lifetime increases (that is, more red pixels) in the GBP–Pon sample. d, GFP lifetime measurements in the samples described in b and c (mean ± s.e.m. of different beads or cells; n, number of beads or cells scored). Note that lifetimes of GFP–Patronin are the same in the spindle and the cortex in the GFP–Patronin + GBP–Pon SOP, and that this homogenous lifetime value in the SOP corresponds to that of (His)6–GFP bound to nanobody beads. This suggests that most GFP–Patronin molecules are bound to the nanobody in the SOP. Please note that in the experiments described in b (purified GFP) and in c (GFP–Patronin fusion in the fly) the term GFP refers to the enhanced GFP variant (eGFP). e, Fluorescence recovery after photobleaching (FRAP) of GFP–Patronin at the spindle. Left panel, images from the FRAP experiment. Fluorescence of GFP–Patronin at the mitotic spindle was bleached (yellow region of interest) and fluorescence recovery was monitored over time. Elapsed time, indicated in seconds. Right panel, average recovery of 11 cells (blue curve; mean ± s.e.m.) was fitted to a single exponential model (red line), providing an estimate of the half-life of GFP–Patronin on microtubules of 1.31 ± 0.03 s (95% confidence interval). This indicates that Patronin exchanges rapidly between the cytosol and the spindle. f–h, Co-expression of GBP–Bazooka and GFP–Patronin does not cause central spindle inversion in the SOP. f, Percentage of iDelta20 in the pIIa daughter cell at late cytokinesis in SOPs expressing GFP–Patronin (GFP–Patronin) or expressing GBP–Bazooka in addition (GFP–Patronin + GBP–Bazooka). Co-expression of GBP–Bazooka with GFP–Patronin does not cause a statistically significant reduction of iDelta20 asymmetry compared to control. (P = 0.847, unpaired two-tailed t-test). g, Dividing SOPs expressing mRFP–Pon and eGFP–Patronin (GFP–Patronin) or expressing GBP–Pon in addition (GFP–Patronin + GBP–Pon) or GBP–Bazooka in addition (GFP–Patronin + GBP–Bazooka) in metaphase. SDCM imaging (single plane). Bottom panels correspond to the rainbow lookup table. GBP–Pon targets GFP–Patronin asymmetrically to the cortex of the pIIb and, conversely, GBP–Bazooka targets GFP–Patronin asymmetrically to the cortex of the pIIa. Note that, in metaphase, the mitotic spindle is symmetric in control, but also in the two GBP conditions. h, Dividing SOPs as in g in late cytokinesis. GBP–Pon inverts the asymmetry of the central spindle, whereas GBP–Bazooka does not (it shows the same asymmetry as in control). SDCM imaging (maximum-intensity projection, except GFP–Patronin in the GFP–Patronin + GBP–Bazooka image, which corresponds to a single plane). n, number of SOPs scored. Scale bars, 5 μm (a, c, e, g, h) and 40 μm (b).
a, Schematic representation of a model of bidirectional motor transport, which generates a biased spatial distribution of endosomes through an antiparallel microtubule overlap (see Supplementary Information equations). b–e, Determination of koff and konρ of endosomes for microtubules and transport run length in control cells (see also ‘Determination of koff and konρ and transport run length’ in Methods for details). b, Representative endosome track at the central spindle (along the x axis, as in Fig. 2e) after automated detection of transport segments (that is, motility occurs by microtubule-based transport; red) and of diffusion segments (that is, movement occurs by diffusion; green). Blue segments correspond to the initial and final segments of the track, the total length of which cannot be determined and are therefore excluded from the analysis. Grey line, extent of the Pavarotti antiparallel overlap of microtubules as in Fig. 2. c–e, Distribution of the duration (c, d) and the run length (e) of transport segments (c, e) and/or diffusion-segments (d) from 101 tracks. Exponential fits (blue lines) give koff = 0.90 ± 0.06 s−1 (c) and kon = 0.05 ± 0.01 s−1 (d; 95% confidence interval). Characteristic run length was not estimated from exponential fit of this distribution in e, but instead by the method of Thorn and Vale (ref. 65), which relies on the fitting of the cumulative distribution of the run length (see Methods). The estimated characteristic transport run length is λ = 0.31 ± 0.01 μm (R2 = 0.98; 95% confidence interval). f, Endosome asymmetric segregation (that is, ; PpIIa, iDelta20 signal in pIIa; PpIIb, in pIIb) as a function of the central spindle asymmetry (that is, ; ρb, GFP–Patronin density in the pIIb side; ρa, in pIIa) for GFP–Patronin control (n = 21) or GFP–Patronin + GBP–Pon (n = 25) cells at during late cytokinesis (∼600 s in registered time). The same data are shown in bins in Fig. 4h. The experimental data points fall on the theoretical curve (grey line corresponding to equation (36), with measured parameters values; see Supplementary Information equations) which accounts for the different levels of spindle reversal and their corresponding endosome segregation reversal measured in this assay. g, Fraction of iDelta20 versus microtubule enrichment as in f, considering the data from all the genotypes presented in this study. The model (grey line, as in f) for asymmetric bidirectional transport accounts for all spindle asymmetry/endosome asymmetries found across all the backgrounds presented in this study. h, Fraction of iDelta20 versus microtubule enrichment as in g. The model outside of the limit case (grey line corresponding to equation (37) with measured parameters values and fitted l = 3.2 μm, see Supplementary Information equations) also accounts for all spindle asymmetry/endosome asymmetries found across all the backgrounds presented in this study.
This file contains uncropped scans with size marker indications for Extended Data Figures 1a, 1f, 1k, 6a. (PDF 393 kb)
This file contains a Supplementary Discussion, Supplementary Equations and additional references. (PDF 710 kb)
Time-lapse spinning disk confocal imaging of representative dividing SOP showing GFP Klp98A, mRFP-Pon to label the pIIb cortex and iDelta20 to label Sara endosomes. Low levels of Klp98-GFP expression were achieved by temperature control. Maximum intensity projection of 25 planes with an increment of 0.5 µm. Klp98A colocalizes with Sara endosomes throughout mitosis, and is therefore asymmetrically segregated to the pIIa daughter cell after cytokinesis. This video corresponds to Fig. 1a. Scale bar: 5 µm. (MOV 517 kb)
Klp98A controls Sara endosome central spindle targeting and asymmetric segregation into the pIIa daughter cell
Time-lapse spinning disk confocal imaging of representative dividing Control and Klp98AΔ47/Δ8 SOPs showing GFP-Pon to label the pIIb cortex, mCherry-Pavarotti to label the central spindle and iDelta20 to label Sara endosomes. In Control cells, Sara endosomes are targeted to the central spindle then are asymmetrically segregated in the pIIa daughter cell. In Klp98AΔ47/Δ8 mutant cells, Sara endosomes neither reach the spindle nor segregate asymmetrically. Maximum intensity projection of 25 planes with an increment of 0.5 µm. This video corresponds to Fig. 1e. Scale bar: 5 µm. (MOV 6353 kb)
Left panel: Spinning-disk confocal imaging of GFP-Pavarotti during SOP division (maximum intensity projection of 7 planes with an increment of 0.5 µm). Middle panel: automatic tracking of the central spindle Cartesian reference frame by tracking of the GFP-Pavarotti signal during SOP division. Right panel: Temporal profile of the GFP-Pavarotti length (PL, x-axis) and width (PW, y-axis) during mitosis. This video corresponds to Fig. 2b and c. Scale bar: 1 µm. (MOV 727 kb)
Top left panel: Time-lapse spinning disk confocal imaging of GFP-Pavarotti and iDelta20 during SOP division (Maximum intensity projection of 7 planes with an increment of 0.5 µm). Bottom left panel: Automated tracking of iDelta20 endosomes during division. Right panel: track of the endosome decomposed on the x-axis in the reference frame of the central spindle. Scale bar: 1 µm. Elapsed time in seconds. (MOV 7239 kb)
TIRFM imaging of the motility of purified Klp98A bound to Quantum dots onto an antiparallel MT overlap composed of Rhodamine GMPPCP-MTs and MAP65-1 in a 0.13:1 molar ratio. Klp98A-coupled Quantum dots (orange arrow) exhibit bidirectional movements within the MAP65-1 positive region. Lower panel corresponds to kymograph along the antiparallel MT overlap. This video corresponds to Fig. 2h. Scale bar: 2 µm. (MOV 1215 kb)
Time lapse average video of 9 SOPs during division. The central spindle is asymmetrical, with an enrichment of Jupiter density on the pIIb side (left). SDCM imaging (maximum intensity projection of 25 planes with an increment of 0.5 µm). Lower right panel corresponds to Rainbow LUT. The central spindle of SOP is asymmetrical, with an enrichment of microtubules on the pIIb (left) side of the spindle. This video corresponds to Fig. 3a and Extended Data Fig. 5. Scale bar: 5 µm. (MOV 824 kb)
Left panel: Time-lapse spinning disk confocal imaging of dividing SOP expressing mRFP-Pon and Jupiter-GFP. Maximum intensity projection of 25 planes with an increment of 0.5 µm. Spindle asymmetry gradually rises during central spindle formation. Middle panel corresponds to Rainbow LUT. Right panel corresponds to average MT enrichment in pIIb over time (Fig. 3e). This video corresponds to Fig. 3. Time stamps in the movies and on the right panel correspond to registered time whose origin is set at the onset of anaphase B (see methods). Scale bar: 5 µm. (MOV 1781 kb)
Time-lapse spinning disk confocal imaging of dividing Control, Klp10A- or Patronin- depleted SOPs showing Jupiter-GFP, mRFP-Pon and Sara endosomes (iDelta20). Maximum intensity projection of 25 planes with an increment of 0.5 µm. The dashed lines represent the pIIa (white) or pIIb (blue) cell membranes. Both central spindle and endosome segregation are symmetric in both Patronin and Klp10A knock-down conditions. This video corresponds to Fig. 4a and b. Scale bar: 5 µm. (MOV 986 kb)
Time-lapse spinning disk confocal imaging of dividing SOP showing mRFP-Pon, iDelta20 and GFP-Patronin (Control), or in addition expressing GBP-Pon (+GBP-Pon), an anti GFP nanobody fused to the anterior cortex localization domain of Pon. Maximum intensity projection of 25 planes with an increment of 0.5 µm. The dashed lines represent the pIIa (white) or pIIb (blue) cell membranes. Expression of GBP-Pon targets GFP-Patronin asymmetrically to the cortex of the pIIb, induces an inversion of central spindle asymmetry (GFP-Patronin channel) and an inversion of the endosome asymmetric segregation (iDelta20 channel). This video corresponds to Fig. 4f,g. Scale bar: 5 µm. (MOV 1923 kb)
Time-lapse spinning disk confocal imaging of dividing SOP showing mRFP-Pon, SiR-tubulin and GFP-Patronin (GFP-Patronin Control), or in addition expressing GBP-Pon (GFP-Patronin + GBP-Pon). Maximum intensity projection of 25 planes with an increment of 0.5 µm. The dashed lines represent the pIIa (white) or pIIb (blue) cell membranes. Expression of GBP-Pon induces an inversion of central spindle asymmetry as seen in the GFP-Patronin and SiR-tubulin images. Scale bar: 5 µm. (MOV 5619 kb)
About this article
Cite this article
Derivery, E., Seum, C., Daeden, A. et al. Polarized endosome dynamics by spindle asymmetry during asymmetric cell division. Nature 528, 280–285 (2015). https://doi.org/10.1038/nature16443
This article is cited by
Polarized branched Actin modulates cortical mechanics to produce unequal-size daughters during asymmetric division
Nature Cell Biology (2023)
Elongator stabilizes microtubules to control central spindle asymmetry and polarized trafficking of cell fate determinants
Nature Cell Biology (2022)
Nature Cell Biology (2022)
Cell Death & Differentiation (2021)