Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 7 (1), 9583

Microtubules Modulate F-actin Dynamics During Neuronal Polarization

Affiliations

Microtubules Modulate F-actin Dynamics During Neuronal Polarization

Bing Zhao et al. Sci Rep.

Abstract

Neuronal polarization is reflected by different dynamics of microtubule and filamentous actin (F-actin). Axonal microtubules are more stable than those in the remaining neurites, while dynamics of F-actin in axonal growth cones clearly exceed those in their dendritic counterparts. However, whether a functional interplay exists between the microtubule network and F-actin dynamics in growing axons and whether this interplay is instrumental for breaking cellular symmetry is currently unknown. Here, we show that an increment on microtubule stability or number of microtubules is associated with increased F-actin dynamics. Moreover, we show that Drebrin E, an F-actin and microtubule plus-end binding protein, mediates this cross talk. Drebrin E segregates preferentially to growth cones with a higher F-actin treadmilling rate, where more microtubule plus-ends are found. Interruption of the interaction of Drebrin E with microtubules decreases F-actin dynamics and arrests neuronal polarization. Collectively the data show that microtubules modulate F-actin dynamics for initial axon extension during neuronal development.

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Figure 1
Figure 1
Drebrin promotes microtubule entry into growth cones. (a–c) Drebrin-YFP expression promotes entry of EB3 comets to the peripheral domain of growth cones compared to growth cones expressing Lifeact-GFP (kymographs from white line). (d) Quantification of growth cone area occupied by EB3 comets. In percentage (%), EB3-mCherry/Lifeact-GFP cells = 52.12 ± 2.182; n = 10 cells from at least three different cultures; EB3-mCherry/Drebrin-YFP cells = 76.51 ± 2.019; n = 10 cells from at least three different cultures; ****p < 0.0001 by t test; Mean ± SEM (e) Representative stage 2 cell transfected with Drebrin-YFP and EB3-mCherry showing the neurite with higher Drebrin intensity receives more EB3 comets. The neurite (n1) that is enriched with overexpressed Drebrin (white arrow) received higher number of EB3 comets (see the kymograph from white lines, n1–n5) (f) Pearson correlation analysis of Drebrin-YFP intensity and EB3 comets number entering into the neurites (per 5 min) from stage 2 cells. n = 8 cells from at least three different cultures; Values were normalized according to standard score and axes are represented in units of standard deviation [σ]. Line equation: Y = 0.6698*X + 5.742e-008; Pearson r = 0.6698, p < 0.0001. Dashed lines represent 95% confidence intervals. Scale bar: 10 μm (e).
Figure 2
Figure 2
Drebrin is segregated to growth cones with higher F-actin treadmilling rates. (a,b) Endogenous Drebrin is enriched in growth cones with higher treadmilling rates. The neurite (white arrow) that has faster - F-actin treadmilling at the growth cone (kymographs n1) accumulates more Drebrin (black arrow in b). (c) Pearson correlation analysis of endogenous Drebrin intensity and F-actin treadmilling speed in growth cones from stage 2 cells. n = 10 cells from at least three different cultures; Values were normalized according to standard score and axes are represented in units of standard deviation [σ]. Line equation: Y = 0.4560*X − 0.009099; Pearson r = 0.4549, p < 0.0004. Dashed lines represent 95% confidence intervals. (d) Stage 2 Drebrin-overexpressed cell segregates preferentially Drebrin-YFP to the growth cones with faster treadmilling (kymographs from white line 1, 2). (e) Second order polynomial fit of Drebrin-YFP intensity and F-actin treadmilling speed in growth cones from stage 2 cells. n = 11 cells from at least three different cultures. Values were normalized according to standard score and axes are represented in units of standard deviation [σ]. Line equation: Y = 0.8723*X − 1.270e-007; R squared value is 0.7740. Dashed lines represent 95% confidence intervals. (f) Early stage 3 Drebrin-overexpressed cell segregates Drebrin-YFP preferentially to the future axon with a more actin-dynamic growth cone (kymographs from white line 1, 2). Scale bar: 10 μm (a,d,f).
Figure 3
Figure 3
Drebrin localization predicts the position of axon outgrowth. (a) Stage 2 and stage 3 neurons transfected with Drebrin-YFP and Lifeact-RFP showing polarized Drebrin signal at one neurite tip. (b) Quantifications showing percentage of stage 2 and stage 3 cells with polarized Drebrin signal. In percentage %, stage 2 = 49.21 ± 6.349, stage 3 = 69.57 ± 6.859, stage 2 cells n = 63, stage 3 cells n = 46, p = 0.0337, Student’s t test, *p < 0.05; Mean ± SEM. (c) Neuron expressing Drebrin-YFP and Lifeact-RFP forms an axon from the dynamic neurite, which shows more Drebrin-YFP (d) Tau-1 staining confirms the axonal identity of the cell. Scale bar: 10 μm (a and c).
Figure 4
Figure 4
Microtubule dynamics predict F-actin treadmilling. (a) F-actin initiated treadmilling in places where more EB3 comets arrived (kymograph at 225°). In places where few EB3 comets were detected F-actin treadmilling is absent (kymograph at 45°). (b) Stage 2 cell shows that the neurite with more EB3 comets (kymographs from white lines) has faster F-actin treadmilling (kymographs from 1–3). (c) Pearson correlation analysis of number of EB3 comets and F-actin treadmilling speed in growth cones from stage 2 cells. n = 9 cells from at least three different cultures. Values were normalized according to standard score and axes are represented in units of standard deviation [σ]. Line equation: Y = 0.6337*X − 0.08005; Pearson r = 0.6395, p < 0.0001. Dashed lines represent 95% confidence intervals. (d) The longest neurite (n1) of the early stage 3 cell that receives more EB3 comets (kymographs from white lines n1–n5) has faster F-actin treadmilling (kymographs from 1–5). Scale bar: 5 μm (a), 10 μm (b and d).
Figure 5
Figure 5
Microtubule dynamics and stability affect growth cone F-actin treadmilling. (a) Neuron transfected with Lifeact-GFP and EB3-mCherry before and after nocodazole treatment. The max projection images show EB3-mCherry signal before and after nocodazole treatment (120 min). Kymographs (from red line 1–4) show progressive reduction of F-actin treadmilling after nocodazole treatment. (b) Neuron expressing Lifeact-GFP before and after taxol treatment. Taxol treatment increased the F-actin dynamics at growth cones (kymographs 1–4 from red lines 1–4). (c) Quantification of F-actin treadmilling in growth cones before and after nocodazole and taxol treatment. Control = 4.7258 ± 0.1918 μm/min; n = 10 cells from at least three different cultures; after nocodazole = 1.6522 ± 0.1183 μm/min; n = 10 cells from at least three different cultures; p < 0.0001 by one-way ANOVA, post hoc Dunnett test ****p < 0.0001; after taxol = 5.7273 ± 0.2150 μm/min; n = 10 cells from at least three different cultures; p < 0.0001 by one-way ANOVA, post hoc Dunnett test ***p < 0.001; Mean ± SEM. (d) Transfection of mouse cortical neurons via in utero electroporation with Cep120 shRNA decreased F-actin treadmilling in growth cones, kymograph drawn from red line 1 and green arrow on the kymograph indicates the F-actin treadmilling slope. (e) Quantification of total neurites, neurites with and without growth cones, from stage 2 transfected with Control shRNA and Cep120 shRNA were shown. Separate statistical comparisons were made to analyze the differences in total neurite number (black + grey bars) and neurites with growth cones (black bars) among the groups. Neurites per cell: Ctrl shRNA = 3.500 ± 0.1593 VS Cep120 shRNA = 2.767 ± 0.1492, **p = 0.0014 by t test, Mean ± SEM. Growth cones per cells: Ctrl shRNA = 1.923 ± 0.1350 VS Cep120 shRNA = 0.3667 ± 0.08949; n = 26 cells for Ctrl shRNA and n = 30 cells for Cep120 shRNA obtained from at least three different cultures; ****p < 0.0001 by t test, Mean ± SEM. (f) Quantification of F-actin treadmilling in growth cones from cells transfected with Control and Cep120-shRNA. Control shRNA = 5.157 ± 0.1927 µm/min, Cep120 shRNA = 2.661 ± 0.1226 µm/min; n = 11 cells per condition, obtained from at least three different cultures; ****p < 0.0001 by t test; Mean ± SEM. Scale bar: 10 μm (a, b and c).
Figure 6
Figure 6
Drebrin knockdown or disruption of microtubule/Drebrin interaction affects F-actin treadmilling and growth cone formation. (a) Drebrin siRNA transfection in rat hippocampal neurons decreased F-actin dynamics in growth cones. Drebrin immunostaining confirms efficient knockdown in Drebrin siRNA transfected neuron pointed with green arrowhead. (b) Quantification of F-actin treadmilling in growth cones from cells transfected with Control siRNA and Drebrin siRNA. Control siRNA = 3.886 ± 0.0922 μm/min; n = 10 cells from at least three different cultures; Drebrin siRNA = 0.321 ± 0.0630 μm/min; n = 10 cells from at least three different cultures; ****p < 0.0001 by t test; Mean ± SEM. (c) Quantification of total neurites, neurites with and without growth cones (gc) from stage 2 and 3 cells transfected with Control siRNA, Drebrin siRNA were shown. Separate statistical comparisons were made to analyze the differences in total neurite number (black + grey bars) and neurites with growth cones (black bars) among the groups. Neurites per cell: Ctrl siRNA = 4.250 ± 0.3286 VS Drebrin siRNA = 2.900 ± 0.4583, Ctrl siRNA n = 12, Drebrin siRNA n = 10 cells from at least three different cultures; *p = 0.0237 by t test. Growth cones per cells: Ctrl siRNA = 1.917 ± 0.2289 VS Drebrin siRNA = 0, Ctrl siRNA n = 12, Drebrin siRNA n = 10 cells from at least three different cultures; ****p < 0.0001 by t test; Mean ± SEM. (d) Expression of EB3 truncation mutants (EB3M and EB3DeltaC) decreased F-actin dynamics in growth cones (kymographs from white lines). (e) Quantification of F-actin treadmilling in growth cones from cells expressing EB3, EB3M, and EB3DeltaC. Control = 5.130 ± 0.1017μm/min; n = 15 cells from at least three different cultures; EB3M = 1.1623 ± 0.0737μm/min; n = 12 cells from at least three different cultures; p < 0.0001 by one-way ANOVA, post hoc Dunnett’s test ****p < 0.0001; EB3DeltaC = 2.1153 ± 0.2027 μm/min; n = 9 cells from at least three different cultures; p < 0.0001 by one-way ANOVA, post hoc Dunnett’s test ****p < 0.0001; Mean ± SEM. (f) Quantification of total neurites, neurites with and without growth cones (gc) from stage 2 and 3 cells transfected with EB3, EB3M, EB3DeltaC were shown. Separate statistical comparisons were made to analyze the differences in total neurite number (black + grey bars) and neurites with growth cones (black bars) among the groups. Neurites per cell, Control EB3 = 5.414 ± 0.3274 VS EB3M = 3.810 ± 0.3128 or EB3DeltaC = 2.081 ± 0.2986; Control EB3 n = 29, EB3M n = 42, EB3DeltaC n = 37 cells from at least three different cultures, p < 0.0001 by one-way ANOVA, post hoc Dunnett’s test, **p < 0.01, ****p < 0.0001. Growth cones per cells, Control EB3 = 1.862 ± 0.1968 VS EB3M = 0.2143 ± 0.0802 or EB3DeltaC = 0.2162 ± 0.07880; Control EB3 n = 29, EB3M n = 42, EB3DeltaC n = 37 cells from at least three different cultures, p < 0.0001 by one-way ANOVA, post hoc Dunnett’s test ****p < 0.0001, **p < 0.01; Mean ± SEM. Scale bar: 5 μm (a) and 10 μm (d).
Figure 7
Figure 7
Drebrin phospho-dead and phospho-mimetic mutants overexpression affects F-actin dynamics and growth cone formation. (a) Expression of Drebrin phospho-dead mutant (DrebrinS142A) decreased F-actin dynamics and Drebrin phospho-mimetic mutant (DrebrinS142D) increased F-actin dynamics in growth cones (kymographs from white lines). (b) Quantification of F-actin treadmilling in growth cones of cells expressing Drebrin, DrebrinS142A and DrebrinS142D. Drebrin = 1.579 ± 0.1030 μm/min; n = 12 cells from at least three different cultures; DrebrinS142A = 1.090 ± 0.0659 μm/min; n = 10 cells from at least three different cultures; DrebrinS142D = 2.08 ± 0.0672 μm/min; n = 10 cells p < 0.0001 by two-way ANOVA, post hoc Tukey’s test ***p < 0.001, ****p < 0.0001; Mean ± SEM. (c) Quantification of total neurites, neurites with growth cones (gc) and growth cones with enriched Drebrin from stage 2 and 3 cells transfected with Drebrin-YFP, DrebrinS142A-YFP, and DrebrinS142D-YFP were shown. Separate statistical comparisons were made to analyze the differences in the number of neurites with growth cone (black + grey bars) and the number of growth cones with enriched Drebrin (black bars) among the groups. Number of Drebrin-enriched growth cones per cell: Drebrin WT = 1.469 ± 0.1626; DrebrinS142D = 2.300 ± 0.4236; DrebrinS142A = 0.6216 ± 0.1362; Drebrin WT n = 49, DrebrinS142D n = 20, DrebrinS142A n = 37 cells from at least three different cultures, p < 0.0001 by one-way ANOVA, post hoc Dunnett’s test, *p < 0.05, **p < 0.01. Total number of neurite (white + black + grey bars): Drebrin WT = 6.551 ± 0.2124 VS DrebrinS142A = 7.486 ± 0.5164 or DrebrinS142D = 5.950 ± 0.4070, Drebrin WT n = 49, DrebrinS142A n = 37, DrebrinS142D n = 20 cells from at least three different cultures, p = 0.0366 by one-way ANOVA, post hoc Dunnett’s test. Number of neurites with growth cone per cell: Drebrin WT = 3.122 ± 0.1812 VS DrebrinS142A = 2.216 ± 0.2424, Drebrin WT n = 49, DrebrinS142A n = 37 cells from at least three different cultures, p = 0.0004 by one-way ANOVA, post hoc Dunnett’s test, **p < 0.01. Mean ± SEM. (de) Confocal images of rat hippocampal cells transfected with Drebrin WT-YFP (d), DrebrinS142A-YFP (e), and DrebrinS142D-YFP (f) were treated with 5 µM Cytochalasin D at DIV2 and fixed with 4% PFA at DIV3 and stained with α-tubulin antibody (shown in red). (g) Quantification of neurite length (µm) after Cytochalasin D treatment. Untransfected control = 42.09 ± 2.813, Drebrin WT = 37.18 ± 3.350, DrebrinS142A = 37.55 ± 2.739, DrebrinS142D = 39.57 ± 2.696, Untransfected control n = 20, Drebrin WT n = 15, DrebrinS142A n = 15, Drebrin S142D n = 20 cells, P = 0.6466 by one-way ANOVA, post hoc Dunnett’s test. Mean ± SEM. Scale bar: 10 μm (a,d,e and f).
Figure 8
Figure 8
Microtubule/Drebrin interaction affects F-actin polymerization in vitro. (a) Kinetics of in vitro reconstitution assay of actin alone and in the presence of, actin with Drebrin or DrebrinS142D and actin, microtubules, EB3, Drebrin or Drebrin S142D. Curves represent mean values obtained from 2–5 independent experiments. (b) Ymax values are calculated from 2–5 independent experiments. Y max values for Actin alone = 1.035 ± 0.04785 A.U.; Actin + Dbn = 0.7504 ± 0.03016; Actin + DbnS142D = 0.8003 ± 0.0116 A.U; Actin + MTs + EB3 + Dbn = 0.6915 ± 0.06875 A.U. Actin + MTs + EB3 + DbnS142D = 0.5267 ± 0.06545 A.U.; *p < 0.05, **p < 0.01, ***p < 0.001 by one-way ANOVA, post hoc Dunnett’s test; Mean ± SEM. (c) Representative TIRF images of in vitro actin polymerization assay under specified conditions. F-actin (labeled by Alexa-Fluor 488 phalloidin) is shown in green and microtubules (MT, labeled by HyLight647) are shown in red. Scale bar: 10 µm.

Similar articles

See all similar articles

Cited by 2 PubMed Central articles

References

    1. Dotti CG, Sullivan CA, Banker GA. The establishment of polarity by hippocampal neurons in culture. J Neurosci. 1988;8:1454–1468. - PMC - PubMed
    1. de Anda FC, Meletis K, Ge X, Rei D, Tsai LH. Centrosome motility is essential for initial axon formation in the neocortex. J Neurosci. 2010;30:10391–10406. doi: 10.1523/JNEUROSCI.0381-10.2010. - DOI - PMC - PubMed
    1. Namba T, et al. Pioneering axons regulate neuronal polarization in the developing cerebral cortex. Neuron. 2014;81:814–829. doi: 10.1016/j.neuron.2013.12.015. - DOI - PubMed
    1. Noctor SC, Martinez-Cerdeno V, Ivic L, Kriegstein AR. Cortical neurons arise in symmetric and asymmetric division zones and migrate through specific phases. Nat Neurosci. 2004;7:136–144. doi: 10.1038/nn1172. - DOI - PubMed
    1. Sakakibara A, et al. Dynamics of centrosome translocation and microtubule organization in neocortical neurons during distinct modes of polarization. Cerebral cortex. 2014;24:1301–1310. doi: 10.1093/cercor/bhs411. - DOI - PubMed

Publication types

Feedback