Figure 1.
Inhibition of Tubulin deacetylases impairs axon elongation.
(A) 2 DIV hippocampal neurons cultured for 48 hours in the presence or absence of TSA (100 nM), and 4 DIV hippocampal neurons treated with TSA from 48 to 96 hours. Neurons were stained with antibodies against MAP2 and Tau-1 to define the somatodendritic and axonal compartments. Scale bar = 50 µm. (B) Percentage of 2 DIV neurons that develop an axon, when treated with TSA (100 or 200 nM) from 5 hours to 48 hours. Data represent the mean ± SEM of 3 independent experiments (300 neurons/experimental condition and experiment). *p<0.05, **p<0.01, paired t-test. (C) Mean axon length of 4 DIV hippocampal neurons treated with TSA (100 or 200 nM) from 48 to 96 hours. Data represent the mean ± SEM of 3 independent experiments (100 neurons/experimental condition and experiment). *p<0.05, paired t-test.
Figure 2.
HDAC6 localizes to axons and it is distributed in the distal region of axons during elongation.
(A) Axonal localization of HDAC6 (green) during axon elongation, where the axon is identified as the tau-1 (red) positive process. Arrows indicate neuronal soma position and arrow heads the axon. Scale bar = 50 µm. (B) Fluorescence intensity along the axons of neurons (a, b and c) shown in A. HDAC6 intensity is shown in colour lines and their respective tau-1 intensities in black lines. (C) Distal region of the axon and growth cone of a 1.5 DIV hippocampal neuron stained for HDAC6 (green), F-actin (red) and acetylated-α-tubulin (blue). Scale bar = 50 µm. (D) 3 DIV neuron treated with cytochalasin D from 1 to 3 DIV and stained for Tau-1 and HDAC6. (E) Western-blot of 3 DIV cultured hippocampal neurons and human HDAC6-Flag transfected N2a cells. Mouse and human HDAC6 were detected using the same antibody that the one used for immunocytochemistry (panel A). Note the difference in molecular weight between human and mouse HDAC6.
Figure 3.
HDAC6 activity inhibition reduces the rate of axon elongation.
(A) α-tubulin acetylation in 2 DIV hippocampal neurons treated with the HDAC6 specific inhibitor tubacin, its non-active analog, niltubacin, or its vehicle, DMSO (control). The graph represents the mean ± SEM of the acetylated-α-tubulin/α-tubulin ratio normalized to control in 3 independent experiments. ****p<0.0001, paired t-test. (B) 2 DIV hippocampal neurons cultured in the presence of vehicle (DMSO), the HDAC6 inhibitor (tubacin, 10 µM) and its inactive compound (niltubacin, 10 µM). Neurons were stained for MAP2 and tau-1. Scale bar = 100 µm. The graph represents the percentage of neurons with a tau-1 positive process. Data represent the mean ± SEM of 3 independent experiments (500 neurons/experimental condition and experiment). ***p<0.001, paired t-test. (C) 4 DIV hippocampal neurons cultured from 48 hours to 4 days in vitro in the presence of DMSO, tubacin (10 µM) or niltubacin (10 µM), and stained as indicated in (B). Scale bar = 100 µm. The graph represents the axonal length of control neurons fixed at 2 DIV or 4 DIV, and of 4 DIV neurons treated from 2 DIV to 4 DIV with tubacin or niltubacin. Data represent the mean ± SEM of 3 independent experiments (150 neurons/experimental condition and experiment). **p<0.01, paired t-test.
Figure 4.
Suppression of HDAC6 by interference shRNAs reduces axonal elongation in hippocampal neurons.
(A) Endogenous HDAC6 in N2a cells is suppressed by the expression of two different HDAC6 shRNAs. Graph represents the mean and SEM of HDAC6 expression levels normalized to β-actin levels in 3 different experiments. ***p<0.001, paired t-test. (B) HDAC6 shRNA suppresses the expression of exogenous HDAC6-Flag in N2a cells and increases acetylated-α-tubulin expression. Graph represents the reduction in HDAC6 expression in N2a cells measured with the anti-Flag antibody. Data represent the mean and SEM of 3 independent experiments. ****p<0.0001, ***p<0.001, paired t-test. (C) 3 DIV hippocampal neurons nucleofected with GFP plasmids expressing HDAC6 interference RNAs or scramble shRNA. Neurons were stained with anti-Tau-1 antibody to identify the axon. Scale bar = 100 µm. (D) Quantification of the mean axonal length of 3 DIV hippocampal neurons nucleofected with the GFP plasmids expressing GFP, scramble shRNA, HDAC6 shRNA 1 or HDAC6 shRNA 2. (E) Box-plot shows the axonal length distribution of neurons quantified in D. ***p<0.001, paired t-test.
Figure 5.
HDAC6 inhibition or suppression interfere the concentration of ankyrinG and voltage dependent sodium channels at the axon initial segment.
(A) Control or TSA treated 6 DIV hippocampal neurons stained with the somatodendritic marker, MAP2 (red), and the AIS markers, ankyrinG or PanNaCh (green). Control neurons concentrate ankyrinG and sodium channels in the AIS, while in TSA treated neurons ankyrinG staining is distributed all along the axon, and voltage gated sodium channels staining is very low or lost. Arrow head indicate the position of the axon. Scale bar = 50 µm. (B) Percentage of neurons with ankyrinG or voltage gated sodium channels concentrated in the AIS. Data represent the mean ± SEM of 3 independent experiments (500 neurons/experimental condition and experiment). ***p<0.001, ****p<0.0001, paired t-test. (C) Western-blot shows the expression levels of the α-subunit of voltage gated sodium channels in control and 100 nM TSA treated neurons compared to β-actin expression levels. (D) 6 DIV hippocampal neurons treated with DMSO, tubacin or niltubacin from 3 DIV to 6 DIV, and then stained for MAP2 (red) and ankyrinG or sodium channels (green). (E) Percentage of neurons that show a localized concentration of ankyrinG or sodium channels at the axon initial segment. The graphs represent the mean ± SEM of 3 independent experiments (600 neurons/experimental condition and experiment). *p<0.05, paired t-test. Scale bar = 100 µm. (F) 3 DIV hippocampal neurons nucleofected with an HDAC6 shRNA plasmid or the control plasmid. After 3 days in culture, neurons were stained for ankyrinG and the neuronal morphology was distinguished by GFP fluorescence. Rectangles indicate the magnified zone shown in lower panels. Lower panels show a magnification of both neurons along the axon with ankyrinG concentrated at the initial region of the axon (control neurons, pGFP-V-RS) or along the axon (pGFP-V-RS-shRNA-HDAC6 nucleofected neurons). Scale bar = 100 µm. (G) 3 DIV hippocampal neurons nucleofected as in E and stained with an anti-PanNaCh antibody. Square inserts show a magnification of both neurons at the level of the AIS. Scale bar = 100 µm. (H, I) Percentage of GFP, scramble shRNA or HDAC6 shRNAs nucleofected neurons that concentrate ankyrinG (G) or sodium channels (H) in the AIS after 3 days in culture. Data are the mean ± SEM of 3 independent experiments (100 neurons/experimental condition and experiment). * p<0.05, **p<0.01, t-test.
Figure 6.
Tubulin deacetylase inhibition disrupts the axon initial segment specific enrichment in detergent resistant acetylated microtubules.
(A) Acetylated-α-tubulin levels in microtubules resistant to PHEM buffer extraction in control and TSA treated neurons. (B) Graph represents the mean ± SEM of α-tubulin acetylation levels from 3 experiments as indicated in A. **p<0.01, *** p<0.001, t-test. (C) Acetylated-α-tubulin expression in 7 DIV hippocampal neurons. Axon initial segment is stained with the pIκBα antibody. (D) 7 DIV hippocampal neurons fixed after extraction with 0.5% Triton X-100 in cytoskeletal buffer (2 mM MgCl2, 10 mM EGTA, 60 mM Pipes pH 7.0) for 5 min at 37°C. Arrows indicate the axon initial segments, stained with acetylated-α-tubulin (red) and pIκBα (green). Box shows an amplification of the indicated axon initial segment. (E) Acetylated-α-tubulin localization in 100 nM TSA treated neurons extracted as indicated in D. Note that acetylated tubulin is not further restricted to the AIS and located along the axon. (F) Acetylated-α-tubulin and total α-tubulin immunostaining of control or 100 nM TSA treated neurons extracted with 0.5% Triton X-100. Arrows indicate the position of neuronal somas. (G) Graph represents the fluorescence intensity of acetylated-α-tubulin (red lines) and α-tubulin (green lines) along the stained axon in control (dotted lines) and TSA (straight lines) treated neurons shown in F. Intensity lines are the result of smoothing the data obtained with the ImageJ program using the Sigmaplot software.
Figure 7.
The absence of HDAC6 activity alters the distribution of KIF5C along the neuron.
(A) KIF5C localization and acetylated-α-tubulin staining in 6 DIV hippocampal neurons cultured in the absence of TSA, or treated with 100 nM TSA from 3 to 6 DIV. Scale bar = 100 µm. (B) Percentage of neurons with a KIF5C with a main axonal localization as shown in A (control neuron). **p<0.01 (t-test). (C) KIF5C localization in 4 DIV hippocampal neurons nucleofected before plating with HDAC6 interference shRNA or scramble. GFP fluorescence indicates the nucleofected neurons. (D) Percentage of scramble or HDAC6 interference shRNA nucleofected neurons with a KIF5C with a main axonal localization. *p<0.05 (t-test). (E) KIF5C localization in 3 DIV HDAC6-GFP or HDAC6 non-active mutant nucleofected neurons. (F) Percentage of neurons showing a KIF5C with a main axonal localization in neurons expressing HDAC6-GFP or an HDAC6-GFP non-active mutant. **p<0.01 (t-test). All data represent the mean ± SEM of three experiments and at least 100 neurons by condition and experiment.