Reduced TUBA1A Tubulin Causes Defects in Trafficking and Impaired Adult Motor Behavior

α-Tubulin protein is decreased in P0 Tuba1a^ND/+ brains. A, Western blottings representing α-tubulin abundance in P0 (left) and adult (right) Tuba1a^ND/+ and wild-type (WT) mouse brains. B, Bar graph quantifying α-tubulin protein in P0 and adult brain tissue lysates relative to GAPDH (N = 3 mice, p < 0.0001 by two-way ANOVA). Quantifications are representative of at least three separate experiments. C, Western blottings showing the abundance of α-tubulin PTMs acetylation (top left), tyrosination (top right), detyrosination (center left), and polyglutamylation (center right), and total α-tubulin (bottom) in adult wild-type and Tuba1a^ND/+ whole-brain lysates. D, Bar graph quantification of Western blotting data shown in C. PTM band volume was normalized to α-tubulin (p > 0.05 for all by t test). E, F, mRNA expression of brain α-tubulin isotypes in Tuba1a^ND/+ and wild-type mice at P0. Data are plotted as fold change in Tuba1a^ND/+ relative to wild type (E) or as a percentage of the total α-tubulin isotype composition (F; p > 0.05 for all by two-way ANOVA). G, H, Bar graph representing mRNA expression of brain α-tubulin isotypes in adult Tuba1a^ND/+ and wild-type mice. Data are plotted as fold change in Tuba1a^ND/+ relative to wild type (G) or as a percentage of the total α-tubulin isotype composition (H; p > 0.05 for all by two-way ANOVA). All Western blotting data represents three replicate blots with lysates from at least three animals per genotype. Bars represent mean ± SEM. ****p < 0.0001.

Reduced TUBA1A function results in fewer microtubule tracks in neurites. A, Schematic of experimental design, illustrating microtubule plus-end tracker, GFP-MACF43, puncta analysis and kymograph generation. B, Representative images of day in vitro (DIV) 1 wild-type (WT) and Tuba1a^ND/+ neurons, visualized using a membrane-bound Myr-TdTomato. Scale bars: 10 μm. C, Representative kymograph plots generated from GFP-MACF43 images in wild-type (left) and Tuba1a^ND/+ (right) neurons. Scale bars: 5 μm. D, Bar graph representing the number of GFP-MACF43 puncta per μm of neurite in DIV1 wild-type and Tuba1a^ND/+ cortical neurons. Bars illustrate mean density of GFP-MACF43 puncta per kymograph and error bars indicate SEM (n = 24 neurons, p < 0.0001). Statistical differences between groups were assessed by t test. ****p < 0.0001.

Tuba1a^ND/+ neurons have deficits in organelle transport caused by increased stationary cargoes. A, Still images from time-lapse microscopy of DIV3 cortical neurons labeled with LysoTracker dye to mark lysosomes in wild-type (WT) and Tuba1a^ND/+ neurons. B, Representative kymograph plots of lysosome movement over time within select neurites for wild type (top) and Tuba1a^ND/+ (bottom). C, Scatter plot representing the percent of stationary lysosomes per kymograph in wild-type and Tuba1a^ND/+ neurons. Data points represent individual neurons (N = 3 mice, n = 25 neurons, p = 0.001). D, Scatter plot representing lysosome velocities in μm/s for wild-type and Tuba1a^ND/+ neurons. Data points represent individual lysosomes (n = 635 lysosomes, p = 0.27). E, Scatter plot representing pause duration (s), for moving lysosomes in wild-type and Tuba1a^ND/+ neurons. Data points represent individual pause events (n = 338 events, p = 0.56). F, Scatter plot representing total distance traveled (μm) by moving lysosomes in wild-type and Tuba1a^ND/+ neurons. Data points represent individual lysosomes (n = 87, p = 0.003). G, Scatter plot representing the percentage of stationary mitochondria in wild-type and Tuba1a^ND/+ neurons. Data points represent individual neurons (N = 3 mice, n = 30 neurons, p = 0.34 by). H, Scatter plot the velocity of short mitochondrial movements in wild-type and Tuba1a^ND/+ neurons. Data points represent individual mitochondria (N = 3 mice, n = 301 mitochondria, p = 0.45). I, Scatter plot representing pause duration (s), for moving mitochondria in wild-type and Tuba1a^ND/+ neurons. Data points represent individual pause events (n = 417 events, p = 0.79). J, Scatter plot representing total distance traveled (μm) by moving mitochondria in wild-type and Tuba1a^ND/+ neurons. Data points represent individual mitochondria (n = 86, p = 0.01). Cortical neurons from three animals per genotype were analyzed. Lysosomes were imaged once every second for 3 min, mitochondria were imaged once every 2 s for 2 min. Statistical differences between groups were assessed by Student’s t test; **p < 0.01; ***p < 0.001. In all plots, line designates mean with error bars indicating SEM. *p < 0.05; **p < 0.01; ***p < 0.001.

Tuba1a^ND/+ mice exhibit adult-onset motor and sensory behavioral deficits. A, Images depicting gait abnormalities in Tuba1a^ND/+ adult mice compared with wild-type (WT). Still images were taken from video of Tuba1a^ND/+ and wild-type mice mid-stride. B, Line graph quantifying changes in rear stance width at two months (p = 0.61), three months (p = 0.02), five months (p = 0.04), eight months (p = 0.03), and 10 months (p < 0.0001) of age in wild-type and Tuba1a^ND/+ mice. C, Line graph representing latency to fall on rotarod task at two months (p = 0.31), three months (p = 0.82), five months (p = 0.03), eight months (p = 0.004), and 10 months (p = 0.04) of age in wild-type and Tuba1a^ND/+ mice. Points depict the mean stance width or latency to fall by genotype with SEM. The same mice were analyzed throughout, and a two-way ANOVA was used to assess statistical significance between groups. Number of mice per genotype is as follows: N = 2 at two months, N = 5 at three months, N = 8 at five months, N = 12 at eight months, N = 8 at 10 months. D, Line graph representing performance over three subsequent rotarod trials in a single testing session at five months of age for Tuba1a^ND/+ compared with wild type. Slopes of the lines for wild type and Tuba1a^ND/+ were compared by linear regression and were not found to be significantly different (p = 0.32, wild-type R² = 0.98, Tuba1a^ND/+ R² = 0.89). E, Scatter plot of grip strength assessed in the forelimbs of female five-month-old wild-type and Tuba1a^ND/+ mice (N = 7, p = 0.19). F, Bar graph of total body weight for eight- to 10-month-old Tuba1a^ND/+ and wild-type female mice (N = 7, p = 0.24). G, Bar graph of SUDO scores Bonin et al. (2014) for Von Frey mechanical sensation testing in three- to six-month-old wild-type and Tuba1a^ND/+ mice (N = 7, p = 0.04). H, Bar graph of time to withdraw from thermal stimulus in Hargreaves behavioral analysis on three-month-old wild-type and Tuba1a^ND/+ mice (N = 7, p = 0.19). Sex as a variable did not significantly impact rear stance width or rotarod performance at any time point examined (N = 7 females, N = 5 males; p > 0.05 for all comparisons of male vs female). Sex was also not found to significantly influence sensory behavior (N = 7 males, N = 2 females; p > 0.05 for all comparisons); *p < 0.05, **p < 0.01, ****p < 0.0001. For line graphs, points indicate mean by genotype at each time point ± SEM. For Bar graphs, bars represent mean ± SEM. *p < 0.05; **p < 0.01; ****p < 0.0001.

Tuba1a^ND/+ does not impact neuronal cell body survival. A, top row, Coronal sections of motor cortex immunolabeled with calbindin (green), ER81 (magenta), and DAPI (blue) in wild-type and Tuba1a^ND/+ mice at three months (left) and 10 months (right) of age at 20× magnification. Center row, Sagittal sections of cerebellum labeled with calbindin (green) and DAPI (blue) in wild-type and Tuba1a^ND/+ mice at three months (left) and 10 months (right) of age at 20× magnification. Bottom row, Nissl-stained coronal sections of the lumbar spinal cord in wild-type and Tuba1a^ND/+ mice at three months (left) and 10 months (right) of age at 4× magnification. B, Scatter plot representing the number of ER81+ Layer V neurons per image in motor cortex at three and 10 months (N = 3 mice, p = 0.19 and p = 0.78, respectively). C, Scatter plot representing the number of calbindin+ Purkinje neurons per image in the cerebellum at three and 10 months (N = 3, p = 0.94 and p = 0.35, respectively). D, Scatter plot representing the number of Nissl+ ventral horn motor neurons per image in the spinal cord at three and 10 months (N = 3, p = 0.66 and p = 0.99, respectively). Motor neurons were identified morphologically. Data points represent ROIs with horizontal line depicting mean with SEM. Data points were nested by animal and analyzed by two-way ANOVA.

No deficits in axon morphology, myelination, or survival in Tuba1a^ND/+ mice. A, Electron micrographs of spinal cord axons in cross-section for adult wild-type (left) and Tuba1a^ND/+ (right) mice. Six images per animal and two animals per genotype were assessed for EM analyses. B, Scatter plot displaying myelin thickness measure (G-ratio) for two wild-type and two Tuba1a^ND/+ mice (N = 2 mice, n = 100 axons, p = 0.42). C, Scatter plot of axon density quantified in electron micrograph images for adult wild-type and Tuba1a^ND/+ mice (N = 2 mice, n = 6 fields, p = 0.72). D, Scatter plot of axon diameter quantified in electron micrograph images for adult wild-type and Tuba1a^ND/+ mice (N = 2 mice, n = 100 axons, p = 0.63). E, Histogram representing the distribution of axon diameters from data in D, F, Anti-SMI-32 (non-phosphorylated neurofilament) immunofluorescence images in cortex of adult wild-type (left) and Tuba1a^ND/+ mice (right); 40× magnification. Scale bar: 10 μm. G, Bar graph representing SMI-32 intensity in a cortical region of interest for wild-type and Tuba1a^ND/+ mice. Data points represent regions of interest, with three animals per genotype assessed (p = 0.77). H, Western blotting for SMI-32 in whole-brain lysates from adult wild-type and Tuba1a^ND/+ mice. Bands for heavy (200 kDa), medium (160 kDa), and light (125 kDa) chain non-phosphorylated neurofilament shown (top). GAPDH used for normalization (37 kDa; bottom). Data points represent technical replicates with three biological replicates performed. I, Bar graphs representing quantification of SMI-32 medium (right, p = 0.61) and heavy (left, p = 0.70) expression in whole-brain lysates, normalized to GAPDH. For all graphs, bar or line indicates mean and error bars display SEM.