MAP1B regulates microtubule dynamics by sequestering EB1/3 in the cytosol of developing neuronal cells
2013; Springer Nature; Volume: 32; Issue: 9 Linguagem: Inglês
10.1038/emboj.2013.76
ISSN1460-2075
AutoresElena Tortosa, Niels Galjart, Jesús Ávila, Carmen Laura Sayas,
Tópico(s)Cellular Mechanics and Interactions
ResumoArticle9 April 2013free access Source Data MAP1B regulates microtubule dynamics by sequestering EB1/3 in the cytosol of developing neuronal cells Elena Tortosa Elena Tortosa Centro de Biología Molecular 'Severo Ochoa' (CSIC-UAM), Department of Molecular Neurobiology, Madrid, Spain Centro de Investigación Biomédica en Red de Enfermedades Neurodegenerativas (CIBERNED), Madrid, SpainPresent address: Division of Cell Biology, Faculty of Science, Utrecht University, Utrecht, The Netherlands. Search for more papers by this author Niels Galjart Niels Galjart Department of Cell Biology and Genetics, Erasmus MC, Rotterdam, The Netherlands Search for more papers by this author Jesús Avila Corresponding Author Jesús Avila Centro de Biología Molecular 'Severo Ochoa' (CSIC-UAM), Department of Molecular Neurobiology, Madrid, Spain Centro de Investigación Biomédica en Red de Enfermedades Neurodegenerativas (CIBERNED), Madrid, Spain Search for more papers by this author Carmen Laura Sayas Corresponding Author Carmen Laura Sayas Centro de Biología Molecular 'Severo Ochoa' (CSIC-UAM), Department of Molecular Neurobiology, Madrid, Spain Centro de Investigación Biomédica en Red de Enfermedades Neurodegenerativas (CIBERNED), Madrid, Spain Search for more papers by this author Elena Tortosa Elena Tortosa Centro de Biología Molecular 'Severo Ochoa' (CSIC-UAM), Department of Molecular Neurobiology, Madrid, Spain Centro de Investigación Biomédica en Red de Enfermedades Neurodegenerativas (CIBERNED), Madrid, SpainPresent address: Division of Cell Biology, Faculty of Science, Utrecht University, Utrecht, The Netherlands. Search for more papers by this author Niels Galjart Niels Galjart Department of Cell Biology and Genetics, Erasmus MC, Rotterdam, The Netherlands Search for more papers by this author Jesús Avila Corresponding Author Jesús Avila Centro de Biología Molecular 'Severo Ochoa' (CSIC-UAM), Department of Molecular Neurobiology, Madrid, Spain Centro de Investigación Biomédica en Red de Enfermedades Neurodegenerativas (CIBERNED), Madrid, Spain Search for more papers by this author Carmen Laura Sayas Corresponding Author Carmen Laura Sayas Centro de Biología Molecular 'Severo Ochoa' (CSIC-UAM), Department of Molecular Neurobiology, Madrid, Spain Centro de Investigación Biomédica en Red de Enfermedades Neurodegenerativas (CIBERNED), Madrid, Spain Search for more papers by this author Author Information Elena Tortosa1,2, Niels Galjart3, Jesús Avila 1,2 and Carmen Laura Sayas 1,2 1Centro de Biología Molecular 'Severo Ochoa' (CSIC-UAM), Department of Molecular Neurobiology, Madrid, Spain 2Centro de Investigación Biomédica en Red de Enfermedades Neurodegenerativas (CIBERNED), Madrid, Spain 3Department of Cell Biology and Genetics, Erasmus MC, Rotterdam, The Netherlands *Corresponding authors. Centro de Biología Molecular, 'Severo Ochoa' (CSIC-UAM), Madrid 28049, Spain. Tel.:+34 91 196 4564; Fax:+34 91 497 4420; E-mail: [email protected] or Tel.:+34 91 196 4592; Fax:+34 91 497 4420; E-mail: [email protected] The EMBO Journal (2013)32:1293-1306https://doi.org/10.1038/emboj.2013.76 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info MAP1B, a structural microtubule (MT)-associated protein highly expressed in developing neurons, plays a key role in neurite and axon extension. However, not all molecular mechanisms by which MAP1B controls MT dynamics during these processes have been revealed. Here, we show that MAP1B interacts directly with EB1 and EB3 (EBs), two core 'microtubule plus-end tracking proteins' (+TIPs), and sequesters them in the cytosol of developing neuronal cells. MAP1B overexpression reduces EBs binding to plus-ends, whereas MAP1B downregulation increases binding of EBs to MTs. These alterations in EBs behaviour lead to changes in MT dynamics, in particular overstabilization and looping, in growth cones of MAP1B-deficient neurons. This contributes to growth cone remodelling and a delay in axon outgrowth. Together, our findings define a new and crucial role of MAP1B as a direct regulator of EBs function and MT dynamics during neurite and axon extension. Our data provide a new layer of MT regulation: a classical MAP, which binds to the MT lattice and not to the end, controls effective concentration of core +TIPs thereby regulating MTs at their plus-ends. Introduction Microtubule (MT) dynamics is crucial during neuronal morphogenesis, allowing rapid morphological changes in neurons in response to extracellular signals. MT dynamics is regulated by different types of MT-associated proteins (MAPs). Among them are classical MAPs, which bind and stabilize MTs along their entire length and MT plus-end tracking proteins (+TIPs) that accumulate specifically at MT growing-ends (reviewed in Amos and Schlieper, 2005). MT-associated protein 1B (MAP1B) is a classical MAP, highly expressed during neuronal development (Tucker et al, 1988, 1989). MAP1B plays important roles during neurite and axon extension, neuronal migration and growth cone navigation, downstream of different signalling pathways (reviewed in Gonzalez-Billault et al, 2004). MAP1B involvement in these processes partly relies on its function as a regulator of MT stability and dynamics. Although MAP1B promotes MT nucleation, polymerization and stabilization, in vitro and in vivo, different studies indicate that MAP1B is a weaker MT stabilizer than two other classical MAPs, tau and MAP2 (Takemura et al, 1992; Pedrotti and Islam, 1995; Vandecandelaere et al, 1996). MAP1B has been shown to mainly bind to dynamic MTs and its ectopic expression in non-neuronal cells (CHO or COS-7) leads to an increase in the population of dynamic MTs (Goold et al, 1999). In developing neurons, MAP1B is prominently located in growing axons, mainly at their distal part and in growth cones, where the proportion of dynamic MTs is very high (Black et al, 1994). Hippocampal neurons from Map1b hypomorphous mice present a reduced proportion of dynamic MTs in the distal part of the axon that correlates with a delay in axon outgrowth (Gonzalez-Billault et al, 2001). In addition, downregulation of MAP1B by RNA interference in cultured cortical neurons leads to slower growing axons and altered MT growth speed in axons (Tymanskyj et al, 2012). It is therefore likely that MAP1B modulates MT dynamics in neurons, but the molecular mechanisms involved are not clear. The end-binding (EB) protein family consists of three members (EB1–3) and is viewed as the 'core' +TIP family (reviewed in Galjart, 2010), since EB1/3 track MT ends autonomously and hence these proteins mark all growing MTs (Lansbergen and Akhmanova, 2006; Bieling et al, 2007, 2008; Dixit et al, 2009; Komarova et al, 2009; Zimniak et al, 2009). Virtually every known +TIP interacts with EB1/3 and many of them require EB1-like proteins for plus-end tracking. In addition, many +TIPs interact with each other at MT plus-ends (reviewed in Galjart, 2010). During neuronal morphogenesis, EB1/3 (as well as other +TIPs) are present in all neuronal compartments, indicating the existence of local MT polymerization throughout the neuron (Stepanova et al, 2003). In differentiating neuroblastoma cells, EB1 regulates MT growth rate, growth distance and duration and its downregulation leads to a reduction in neurite length (Stepanova et al, 2010). Of the three family members, EB3 is predominantly expressed in brain, in particular in neurons (Nakagawa et al, 2000). EB3 is enriched in growth cones and is involved in the coordination of the interaction between F-actin and dynamic MTs during neuritogenesis (Geraldo et al, 2008). Hence, EBs (EB1/3) function as local regulators of MT dynamics during neuronal development. We hypothesized that MAP1B and EB1/3 might act in a cooperative manner to regulate MT dynamics during neurite and axon outgrowth. Our results show that overexpression of MAP1B in neuroblastoma cells results in decreased binding of EBs to MT plus-ends. Reciprocally, MAP1B knockdown increases EB1/3 binding to MT growing-ends in correlation with an increase in MT growth speed. Immunofluorescence analyses, co-immunoprecipitation, pull-down and FRAP assays reveal that MAP1B interacts with EBs and sequesters these +TIPs in the cytosol. We provide evidence for an enhanced binding of EB1/3 to MTs and an altered EB3 behaviour in axons and growth cones of MAP1B-deficient neurons. This is reflected in changes in MT growth speed and direction, as well as an increase in MT pausing and looping, which correlate with a delay in axon outgrowth. In summary, we provide molecular insight into how MAP1B regulates locally MT dynamics during neuronal development via its direct interaction with EB1 and EB3 proteins in the cytosol and how this contributes to proper neurite/axon extension. Results MAP1B and EB1/3 localize in neurites and growth cones of differentiating neuronal cells We started analysing the localization of MAP1B and EB1/3 in differentiating mouse neuroblastoma N1E-115 cells, which flatten and elongate neurites upon serum withdrawal. Confocal pictures showed that MAP1B and EB1/3 localized prominently in extending neurites and growth cones (Figures 1A and B). As seen in flat cells, MAP1B localized along the lattice of dynamic (tyrosinated) MTs, whereas EBs accumulated in comet-like dashes at MT plus-ends (Figures 1C and D). These results show that MAP1B and EBs (EB1 and EB3) are enriched in elongating neurites and growth cones and are present at growing (dynamic) MTs in differentiating neuronal cells. Figure 1.Localization of MAP1B and EB1/3 in serum-starved N1E-115 neuroblastoma cells. Confocal images of differentiating N1E-115 cells, either bearing neurites (A, B) or flat (C, D). In (A, C), cells were co-stained with anti-MAP1B (N-t, red) and anti-EB1 (green) and in (B, D), cells were triple stained with anti-MAP1B (N-t, red), anti-EB3 (green) and anti-α-Tyrosinated-tubulin (blue). Scale bars=10 μm. Download figure Download PowerPoint Overexpression of MAP1B results in impaired binding of EBs to MT plus-ends To examine the function of MAP1B on MT dynamics and neurite outgrowth, we first analysed localization of EB proteins in differentiating neuroblastoma cells ectopically expressing GFP-tagged MAP1B. MAP1B-GFP overexpression at low to medium levels led to a significant reduction (∼2-fold) in EB1 comet number (Figures 2A and B). Remaining comets were substantially reduced in length (∼2-fold) (Figures 2A and C) and in overall fluorescence intensity, as compared to comets in control cells (Figure 2D), indicating that EB1 binding to MT plus-ends was impaired by an excess of MAP1B. Remarkably, overexpression of GFP-tagged tau—another classical neuronal MAP—at low to medium levels, did not alter significantly EB1 localization at MT plus-ends (Supplementary Figures S1A–C). Although MAP1B-GFP and GFP-tau localized both in the cytosol and along the MT lattice, cytosolic localization was more prominent for MAP1B-GFP (Figure 2A; Supplementary Figure S1A). Moreover, ectopic expression of moderate levels of MAP1B or tau in N1E-115 cells did not significantly increase MT numbers or bundling (Figures 2A, E and F; Supplementary Figure S1D). Expression of GFP-tau led to a slight rise in MT stabilization (Supplementary Figure S1D). These data show that MAP1B overexpression displaces EB1 from MT plus-ends and indicate that this effect is specific for MAP1B and is not mediated by its actions on MTs. These results suggest a novel phenomenon, namely that crosstalk exists between a classical MAP and a 'core' +TIP in neuronal cells. Figure 2.MAP1B overexpression displaces EBs from MT plus-ends. (A) Confocal pictures of N1E-115 cells transfected with MAP1B-GFP (green) and stained with an anti-EB1 antibody (red). Scale bar=10 μm. Insets show details of EB1 comets in a control (non-transfected) (a) versus an MAP1B-GFP transfected cell (a′). (B) EB1 comets per 100 μm2 in control cells and cells transfected with MAP1B-GFP. (C) Average length of EB1 comets. Error bars in (B, C) are s.e.m. Number of comets measured was n=419 (control) and n=415 (MAP1B-GFP transfected). (D) Average fluorescence intensity profiles of EB1 dashes in control and MAP1B-GFP-transfected cells. (E) Confocal images of N1E-115 cells transfected with MAP1B-GFP (green). Total MTs were visualized with anti-α-tubulin (red) and stable MTs with anti-α-Acetylated-tubulin (blue). (F) Quantification of the average fluorescence intensity (±s.e.m.) of the whole MT network (α-tubulin) and of stable MTs (α-Acetylated-tubulin) in control and MAP1B-GFP-expressing cells. No significant changes in MT density or stability were observed in transfected cells. **P<0.005. Download figure Download PowerPoint MAP1B downregulation enhances binding of EBs to MTs To confirm whether MAP1B interplays with EB1/3 in neuronal cells, we knocked down MAP1B expression and analysed EB1/3 localization in MAP1B-depleted cells. Different shRNA constructs were used to generate N1E-115 cell lines stably depleted of MAP1B. A control cell line was generated by using a scramble shRNA construct. MAP1B levels were reduced to different extents, as verified by immunofluorescence and western blot analysis (Figures 3A and B). EB1 and EB3 levels remained unaffected (Figure 3B). MAP1B-deficient cells presented a significantly reduced amount of EB comets (Figure 3C) in correlation with an enhanced accumulation of EB1/3 at MT plus-ends (Figures 3D–F and H) and a diminished MT density (Figures 3E and G). Moreover, EB1/3 partly localized on MT segments (Figures 3E and H), which, due to their length and the equal fluorescence intensity distribution of EB1/3 along the segments, were clearly distinct from MT comets. A similar increased interaction of EB1 with the MT lattice was found in cells overexpressing medium to high levels of EB1 (Supplementary Figures S2A–C). Thus, MAP1B knockdown leads to a reduction in MT density along with a decrease in EBs comet number and a prominent increase in the binding of EB proteins to MT growing-ends and to the MT lattice. Remarkably, interaction of EB1/3 with MTs was not enhanced in tau-depleted cells (Figure 3H; Supplementary Figures S3A and B), which also presented a reduction in MT density (Figure 3G; Supplementary Figure S3C), indicating that the effect of MAP1B on EBs localization is specific, direct and not due to its action on the MT lattice. Collectively, these results indicate that the interaction of EBs with MTs is—at least partly—dependent on MAP1B levels. Figure 3.MAP1B stable downregulation enhances binding of EB1/3 to MTs. N1E-115 stable cell lines generated by lentiviral infection with either scramble shRNA (control) or different specific MAP1B-shRNAs (1, 2, or a pool of 1+2+3). Reduction in MAP1B levels was confirmed by immunocytochemistry (A) (anti-MAP1B (N-t, red) and anti β3-tubulin (green)) and western blot (B), in which MAP1B, EB1 and EB3 levels were analysed and β-actin was used as a loading control. (C) Quantification of average number of EB1 comets/100 μm2 (±s.e.m.) in control and MAP1B-knocked down cells. (D) Average comet length (±s.e.m.) of EB1 or EB3-positive MT tails in control and MAP1B-deficient cells. Number of comets measured in each experimental condition was (a) EB1 staining; n=877 (control), n=1176 (shRNA pool); (b) EB3 staining; n=1497 (control), n=1165 (shRNA pool). Number of cells counted were n=26 (control) and n=55 (shRNA pool). (E) Confocal immunofluorescence pictures of control and MAP1B-silenced cells, co-stained with anti-EB1 (green) and anti-α-tubulin (red). Scale bar=10 μm. Details are presented in insets. (F) Average fluorescence intensity profile of EB1 dashes in control and MAP1B-depleted cells. EB1 interaction is enhanced upon MAP1B knockdown. (G) Quantification of the fluorescence intensity of the MT network (α-tubulin) (±s.e.m.) in control cells and in cells deficient in either MAP1B or tau. (H) Confocal pictures of control (scramble), MAP1B-depleted and tau-depleted cells, stained with anti-EB3. EB3-positive comets and segments are shown at higher magnification in insets. Scale bar=10 μm. Arrowheads in (E, H) point to MT segments highlighted by EB1 (E) or EB3 (H). ***P<0.0005.Source data for this figure is available on the online supplementary information page. Source Data for Figure 3b [embj201376-sup-0001-SourceData-S1.pdf] Download figure Download PowerPoint MAP1B interacts directly with EB1 and EB3 Since our results indicate that the localization of EB1/3 is regulated by MAP1B, we addressed whether MAP1B interacted with these proteins. To test this hypothesis, we first overexpressed EB1 or EB3 fused to GFP (with the tag located either at the N- or at the C-terminus of the proteins) in N1E-115 cells, which were then differentiated by serum starvation, and performed co-immunoprecipitation (co-IP) assays using an antibody against the N-terminal region of MAP1B (Figure 4A). GFP-tagged EB proteins but not GFP, which was used as a negative control, did co-IP with MAP1B. Although expression levels of all GFP-tagged EB constructs were comparable (see input lanes in Figure 4A), clear differences in pull-down efficiencies of EBs with MAP1B were found. In fact, regardless of the position of the GFP tag, EB3 was consistently more abundant in the complex with MAP1B than EB1 (Figure 4A). N-terminal tagging of GFP significantly impairs EBs accumulation at MT plus-ends, whereas the C-terminal tag alters EBs binding to protein partners containing a CAP-GLY domain (such as CLIP-170) (Skube et al, 2010). In our assays, although both GFP-tagged versions of EBs did co-IP with MAP1B, GFP-EBs co-IPed better with MAP1B than EBs-GFP. This suggests that either the interaction is favoured when EBs are not localized at MT plus-ends or that the C-terminal tag partly impairs binding of EB1/3 with MAP1B. Figure 4.MAP1B interacts directly with EB1 and EB3. (A) N1E-115 cells were transfected with GFP (control) or GFP-tagged EB1 or EB3, either at their N- or C-terminal regions. Cells were serum starved overnight and co-IP assays were performed with an antibody against MAP1B (N-t). Expression of each construct was confirmed by western blot using anti-GFP (Inputs), and IP of MAP1B was corroborated with anti-MAP1B (lower blot). Co-IP of GFP-tagged EBs with MAP1B in control or Nocodazole-treated cells (10 μM, 20 min) was confirmed by western blot using an anti-GFP antibody. (B) Co-IP of endogenous MAP1B and EB3 proteins from E18 mouse brain lysates. (C) MAP1B from E18 mouse brain lysates was pulled down with GST-EB1 and GST-EB3 but not with GST (control). (D) In vitro pull-down assays of MAP1B-N-t (1–508)-6x-His with either GST or GST-EBs. The MAP1B-N-t (1–508) fragment interacts directly with GST-EBs but not with GST. Expression of each construct was confirmed and shown by Ponceau staining of the nitrocellulose membrane (C) or by Coomassie staining of the acrylamide gel (D) (n=3 in each case).Source data for this figure is available on the online supplementary information page. Source Data for Figure 4a [embj201376-sup-0002-SourceData-S2.pdf] Source Data for Figure 4c [embj201376-sup-0003-SourceData-S3.pdf] Source Data for Figure 4d [embj201376-sup-0004-SourceData-S4.pdf] Download figure Download PowerPoint To test whether the interaction between MAP1B and EB1/3 was MT dependent, we treated neuroblastoma cells overexpressing GFP-tagged EBs with Nocodazole (10 μM, 20 min) to induce complete MT depolymerization and release of EBs and MAP1B into the cytosol (Supplementary Figures S4A–C). Co-IP assays (Figure 4A) and confocal microscopy (Supplementary Figure S4C) revealed enhanced interaction between EB3 and MAP1B in Nocodazole-treated cells. Thus, formation of the MAP1B/EB protein complex does not require MT integrity and is favoured when both proteins are present in the cytoplasm. Moreover, endogenous EB3 and MAP1B co-IPed in lysates from E18 mouse brains (Figure 4B), indicating that the complex MAP1B/EB3 occurs in vivo. In addition, MAP1B from E18 mouse brain extracts was pulled-down with both GST-EB1 and GST-EB3 but not with GST alone (Figure 4C). Interaction of MAP1B with MTs is regulated by phosphorylation by different kinases, such as proline-directed kinases including glycogen synthase kinase-3 (GSK-3) and cyclin-dependent kinase-5 (Cdk5) (Garcia-Perez et al, 1998; Goold et al, 1999), and non proline-directed kinases such as casein kinase 2 (CK2) (Ulloa et al, 1993). We tested whether the formation of the MAP1B/EB3 complex was regulated by phosphorylation by any of these kinases. Co-IP assays showed that binding of EB3-GFP to MAP1B was enhanced in cells treated with inhibitors of GSK-3 (Lithium, 20 mM, 3 h) or Cdk5 (Roscovitine, 20 μM, 3 h) but was not altered in cells treated with a CK2 inhibitor (DMAT, 10 μM, 3 h) (Supplementary Figure S5A). The interaction between endogenous MAP1B and EB3 was also increased upon Cdk5 inhibition (Supplementary Figure S5B). These results indicate that binding of MAP1B to EB3 is regulated by phosphorylation mediated by proline-directed kinases. We then examined whether the interaction between MAP1B and EBs was direct, by performing in vitro pull-down assays using recombinant GST-EBs and an N-terminal fragment of MAP1B (aa 1–508) with a 6x-His tag. Pull-down assays using glutathione beads showed that (this fragment of) MAP1B interacted directly with both EB1 and EB3 (Figure 4D). Taken together, these results indicate that EB1/3 and MAP1B directly interact and are found in a complex in the cytosol of developing neuronal cells. Although MAP1B interacts with both EB1 and EB3, binding with EB3 is consistently more prominent than with EB1 in cells, in embryonic brain and in vitro. MAP1B sequesters EBs in the cytosol Our results suggest that MAP1B might control the localization of EB proteins through their sequestration in the cytosol. To test this hypothesis, we analysed whether overexpression of N-terminal MAP1B (aa 1–508) mimicked the effects of full-length MAP1B on EB1 localization. MAP1B 1–508 does not contain the MT binding domain. (Togel et al, 1998) and as shown above this MAP1B fragment interacts directly with EB proteins (Figure 4D). Ectopically expressed MAP1B 1–508-Myc showed a diffuse cytoplasmic distribution (Supplementary Figures S6A and B) and had no effect on MT density (Supplementary Figures S6B and C). However, cells expressing Myc-tagged MAP1B 1–508 presented a prominent diffuse cytosolic localization of EB1, concomitant with a significant reduction in both the number and length of EB1 comets (Supplementary Figures S6A, D and E), similar to the decrease found in cells overexpressing full-length Myc-tagged MAP1B (Supplementary Figures S6D and E). These results indicate that EB proteins are sequestered by MAP1B in the cytosol. EB3 cellular mobility is modulated by MAP1B The dynamic behaviour of EB1-like proteins is determined by diffusion and a combination of protein–protein interactions, including fast exchange at MT plus-ends (Dragestein et al, 2008). Since MAP1B preferentially interacts with EB3, we focused on EB3 in the rest of our study. To check whether the direct interaction with MAP1B affects EB3 cellular dynamics during neurite extension, we performed fluorescence recovery after photobleaching (FRAP) assays. Bleaching was done in the distal part of extending neurites of differentiating control or MAP1B-deficient neuroblastoma cells, expressing low levels of EB3-GFP. Since binding reactions with other proteins affect the slow phase of the fluorescence recovery curve, we focused on this region in the FRAP analysis. Interestingly, EB3-GFP fluorescence recovery was faster in MAP1B-depleted cells than in control cells (kcontrol-Vehicle=0.1172±0.0393s−1; kMAP1B-sh-RNA pool-Vehicle=0.1954±0.0531s−1 and t1/2-control-Vehicle=11.70±3.335 s; t1/2-MAP1B-shRNA pool-Vehicle=7.7078±2.079 s) (Figure 5A, Vehicle; t1/2=ln(2)/K). These results indicate that EB3 mobility is increased in the absence of MAP1B. Consistent with these findings, MAP1B-GFP overexpression had opposite effects, with a slight retardation in fluorescence recovery of EB3-mCherry (Supplementary Figure S7). These results indicate that the interaction between EB3 and MAP1B in the distal part of growing neurites results in lowered EB3 mobility. Figure 5.EB3 mobility in distal neurites is regulated by MAP1B. Analysis of slow FRAP in control (scramble) and stably MAP1B-depleted (MAP1B-shRNA pool) N1E-115 cells. Cells were transfected with EB3-GFP, serum starved overnight, and either treated with vehicle (DMSO 0.1%, 1 h) or Nocodazole (10 nM, 1 h) and subjected to FRAP in distal neurites. Data collected from three different sets of experiments (∼8–15 cells/experiment) were normalized and fitted with a two-phase association equation. (A) k (1/s) corresponding to the slow phase of fluorescence recovery in vehicle-treated and Nocodazole-treated cells. (B) Curves of averaged actual FRAP data in control and MAP1B-depleted cells, upon Nocodazole treatment. (C) Representative example of time lapses of FRAP in distal neurites of control and MAP1B-silenced cells, treated with Nocodazole. *P<0.05. Download figure Download PowerPoint Because our biochemical results indicated that the interaction between MAP1B and EBs occurs preferentially in the cytosol, we assessed EB3-GFP dynamics in the presence of Nocodazole. In this experiment, Nocodazole was used at low concentrations (10 nM) to avoid neurite retraction. At this concentration, Nocodazole altered MT dynamics, inducing the removal of most EB3 from MT growing-ends and releasing it in the cytoplasm, as described (Jaworski et al, 2009). Addition of Nocodazole increased EB3-GFP mobility, both in control and in MAP1B-knocked down cells, as shown by an increased k (kControl-Noc.=0.1463±0.0328s−1 and kMAP1B-shRNA pool-Noc.=0.3025±0.0650s−1) and a reduced half-time of fluorescence recovery (t1/2-Control-Noc.=7.7047±1.578 s and t1/2-MAP1B-shRNA pool-Noc.=3.578±0.6378, s) (Figure 5A, Nocodazole, and Figure 5B). Remarkably, Nocodazole enhanced the effect of lack of MAP1B on EB3 dynamics, accelerating fluorescence recovery of EB3-GFP further (Figures 5A–C; Supplementary Figure S8). Taken together, these results confirm that the interaction between MAP1B and EB3 is functional and that MAP1B controls the dynamics/mobility of cytosolic EB3. Moreover, the fact that Nocodazole treatment does not inhibit but exacerbates the effect of MAP1B deficiency on EB3 mobility, supports our biochemical findings showing that complex formation between MAP1B and EB3 occurs in the cytosol. MAP1B regulates EB3 dynamics at MT plus-ends Our results indicate that in differentiating neuronal cells, EB3 mobility and binding to MT plus-ends are negatively regulated by MAP1B. To analyse whether MAP1B controlled MT dynamics at plus-ends, we performed time-lapse confocal experiments in control and MAP1B-depleted N1E-115 cells, transfected with EB3-GFP. EB3-GFP comet displacements (and therefore growing MTs) were manually tracked in living cells. In agreement with the results obtained with endogenous proteins in fixed cells, living cells depleted of MAP1B, presented longer EB3 dashes than control ones (snapshots and insets in Figure 6A and quantification in Figure 6B; Supplementary Movies S1 and S2). MAP1B-silenced cells also displayed a reduction in the density of growing MTs (snapshots and time-lapse insets in Figure 6A and Supplementary Movies S1 and S2), as evidenced by the decrease in the number of EB3-GFP comets per 100 μm2 (Figure 6C). This effect was more prominent close to cell edges, which presented a high concentration of EB3 dashes in control cells but not in depleted cells (snapshots, Z-max projections and insets in Figure 6A). Figure 6.MAP1B regulates EB3 dynamics at MT plus-ends. EB3-GFP was transfected into control (scramble) or MAP1B-depleted (with shRNA2 or a pool of shRNAs 1–3) neuroblastoma cells. Cells were serum starved overnight and EB3-GFP displacements were analysed. Examples of still images of control and knocked down cells are shown in (A) (Snapshot). Pictures were taken every 2 s for 30 frames. Z-maximal projections of the whole time-lapse series are shown in (A). Scale bar=10 μm. Details are shown in insets. A fragment of a representative time lapse of EB3-GFP-comet tracking (0–16 s) is shown for control and MAP1B-depleted cells in small insets under the main pictures (A). Arrows follow displacements of individual comets in the selected frames. EB3-GFP comets were manually tracked and different parameters were measured in control and MAP1B-silenced cells. (B) EB3-GFP comet length (μm). (C) Number of comets per 100 μm2. In (B, C), number of comets measured was n=382 (control), n=303 (MAP1B-shRNA2) and n=150 (MAP1B-shRNA pool). (D) MT growth speed (μm/s). Number of EB3-GFP comets tracked in each case was n=187 (control), n=189 (MAP1B-shRNA2) and n=77 comets (MAP1B-shRNApool). (E) Duration of MT growth events (s). Average values±s.e.m. are represented. *P<0.05, **P<0.005, ***P<0.0005. Download figure Download PowerPoint Moreover, MT plus-end tracking analysis indicated that MAP1B silencing induced a significant increase in MT growth speed (inset
Referência(s)