Proliferating versus differentiating stem and cancer cells exhibit distinct midbody-release behaviour

ARTICLE Received 26 Jul 2011 | Accepted 14 Sep 2011 | Published 18 Oct 2011 DOI: 10.1038/ncomms1511 Proliferating versus differentiating stem and cancer cells exhibit distinct midbody-release behaviour Andreas W. Ettinger1, Michaela Wilsch-Bräuninger1,*, Anne-Marie Marzesco1,*, Marc Bickle1, Annett Lohmann1, Zoltan Maliga1, Jana Karbanová2, Denis Corbeil2, Anthony A. Hyman1 & Wieland B. Huttner1 The central portion of the midbody, a cytoplasmic bridge between nascent daughter cells at the end of cell division, has generally been thought to be retained by one of the daughter cells, but has, recently, also been shown to be released into the extracellular space. The significance of midbody-retention versus -release is unknown. Here we show, by quantitatively analysing midbody-fate in various cell lines under different growth conditions, that the extent of midbody-release is significantly greater in stem cells than cancer-derived cells. Induction of cell differentiation is accompanied by an increase in midbody-release. Knockdown of the endosomal sorting complex required for transport family members, Alix and tumour-suppressor gene 101, or of their interaction partner, centrosomal protein 55, impairs midbody-release, suggesting mechanistic similarities to abscission. Cells with such impaired midbody-release exhibit enhanced responsiveness to a differentiation stimulus. Taken together, midbody-release emerges as a characteristic feature of cells capable of differentiation. Max Planck Institute of Molecular Cell Biology and Genetics, Pfotenhauerstr. 108, Dresden 01307, Germany. 2 BIOTEC, Technische Universität Dresden, Tatzberg 47-49, Dresden 01307, Germany. *These authors contributed equally to this work. Correspondence and requests for materials should be addressed to W.B.H. (email: ). 1 nature communications | 2:503 | DOI: 10.1038/ncomms1511 | www.nature.com/naturecommunications © 2011 Macmillan Publishers Limited. All rights reserved.  ARTICLE nature communications | DOI: 10.1038/ncomms1511 T he midbody is a transient structure formed during cytokinesis in animal cells by the ingression of the cleavage furrow1–4. It constitutes a cytoplasmic bridge between the two nascent daughter cells that contains the remnants of the central spindle and the contractile ring5,6. The central part of the midbody, also called Flemming body, is characterized by a morphologically distinct, electron-dense matrix7–9. Abscission, which completes cell division, severs the midbody-bridge, bypassing its central matrix. Consequently, the post-abscission midbody with its characteristic matrix is inherited asymmetrically by one of the daughter cells10–12. Until recently, the canonical view was that the post-abscission midbody remains associated with the cell that inherited it4,11–13. This view has largely been based on the observation of intracellular midbodies in cells in culture, often referred to as ‘midbody rings’ because of their appearance on analysis of certain midbody markers11,14,15. These internalized post-abscission midbody structures are eventually degraded by autophagy15. However, studies on the neuroepithelium in vivo have recently revealed an alternative fate of the midbody after cell division, that is, its release into the extracellular fluid16–18. This post-abscission route of the midbody constitutes an irreversible disposal of both, the cytoplasmic and membraneous midbody components from the cell, which is not necessarily true in the case of intracellular degradation. In this context, it is noteworthy that not only components of the cytoskeleton such as central spindle and contractile ring constituents, but also certain membrane components become concentrated at the midbody during cell division. For example, the fivetransmembrane-domain protein prominin-1 (CD133)19–21, a marker L929 97 191 191 97 97 97 191 * * * * * 250 148 * H SC 8N S N S− ES 5 SH -R1 -S N Y5 eu Y ro M 2a C F− 7 M P-1 C 9 F− 10 A 3T 3 C H O H H eLa EK 29 3 L9 29 Stem cells h i Six-well 10-cm C P1 P2 P3 P4 Cancer/immortalized cells C P1 P2 P3 P4 (kDa) CRIK 80 Proportion loaded (%) 2 33 33 33 33 0.4 20 20 20 20 60 j 40 20 * 0 N 180 180 * G 191 SH-SY5Y CGR8-NS 116 97 191 * R 116 180 * Stem cells * * * * * * * * 180 10 8 6 4 2 0 xw 10 ell -c m MCF-7 0 180 C P1 P2 P3 (kDa) Si MCF-10A 116 180 * MKLP-1 C P1 P2 P3 (kDa) HSC 10 f CRIK 20 ro M 2a C F− 7 M P-1 C F− 9 10 A 3T 3 C H O H H eLa EK 29 3 L9 29 HEK293 116 e 30 eu CHO 180 116 250 HSC MKLP-1 C P1 P2 P3 (kDa) 148 40 N S− ES 5 -R 1 3T3 180 50 C P-19 180 d CRIK C P1 P2 P3 (kDa) SC ES-R1 116 c 60 H HeLa C P1 P2 P3 P4 (kDa) MKLP-1 in P1, P2, P3 (% of total) Neuro-2a g MKLP-1 C P1 P2 P3 P4 (kDa) NS-5 Results Midbody-release differs between cell lines. To explore how widespread a phenomenon midbody-release is, we subjected the conditioned medium of various cell lines (Supplementary Table S1) to a previously established differential centrifugation protocol16, and analysed the resulting four pellets (P1–P4) by immunoblotting for markers of the central region of the midbody, citron rho-interacting kinase (CRIK)26,27 and mitotic kinesin-like protein-1 (MKLP-1)28 (Fig. 1). In the case of the mouse neural stem cell lines NS-5 and CRIK in P1-4 (% of total) b CRIK CRIK in P1, P2, P3 (% of total) a of many somatic stem cells and cancer stem cells22–24, is clustered at the central part of the midbody-bridge in neuroepithelial and haematopoietic stem and progenitor cells17,25, and has been shown to be released from neuroepithelial cells along with midbodies at the onset of neurogenesis in vivo16,17. Thus, the possibility arises that midbody-release may be associated with changes in cell fate. The existence of midbody-release as an alternative pathway to intracellular degradation of post-abscission midbodies raises three key questions. First, how is midbody-release related to abscission? Second, how do cell types that dispose of midbodies predominantly by release differ from cell types that predominantly degrade midbodies intracellularly? Third, what is the functional significance of midbody-release? Here we have addressed these questions by studying midbody-release versus -retention in various cell types and growth conditions in vitro. We find that midbody-release was greater in stem than cancer cells, and that induction of differentiation enhanced midbody release, and we conclude that midbody release may be a feature of cells that have the ability to commit to differentiation. Cancer/immortalized cells Figure 1 | Release of membrane particles containing midbody markers from various cell types. Cell lines were grown to subconfluency over a period corresponding to two doubling times, and HSCs were grown for 5 days on MSCs, to condition the respective culture medium. Conditioned media were subjected to differential centrifugation to obtain the P1, P2, P3 and P4 (e (...truncated)