Mitochondrial distribution in H4 cells is altered by tau expression. Representative photomicrographs (a) show H4 cells transfected with either GFP or GFP-Tau 4 repeat plus 2 plus 3 lacking the last 20 aa (Tau4RC3). Cells were co-transfected with GFP or GFP-Tau4RC3 (green) and Mito-DsRed (red) to label mitochondria demonstrating the clumping of mitochondria around the nucleus in Tau-transfected cells. Quantification of mitochondrial distribution (b) shows a significant reduction in the area of cell soma occupied by mitochondria with overexpression of tau (# p < 0.0001 Mann–Whitney test tau vs. control). Four-repeat tau isoforms had a significantly more pronounced effect on mitochondrial clumping (* Mann–Whitney test 3 repeat vs. 4 repeat p < 0.0001). Within the 3 repeat and 4 repeat groups, addition of the N-terminal inserts or truncation at the caspase cleavage site (C3) did not have a further effect on mitochondrial distribution. The data are non-parametric thus are presented as box plots, which display the median value (line inside the box), upper quartile (top of the box), lower quartile (bottom of the box), 90th percentile (top whisker), 10th percentile (bottom whisker), and all values below the 10th and above the 90th percentile (potential outliers) shown as dots. n = 45 cells per condition. Scale bar 20 μm.

Mitochondrial distribution in neurons plated on microfluidic devices is altered by expression of human tau. In neuronal cell bodies (a), expression of tau significantly decreased the area of the cell body occupied by mitochondria (# p < 0.0001, Mann–Whitney test control vs. tau). Four-repeat isoforms had a more pronounced effect than 3 repeat isoforms (* p < 0.0001, Mann–Whitney test 3 repeat vs. 4 repeat, tau group). Within the 4 repeat isoforms, the shortest 4 repeat construct, 4 repeat minus 2 minus 3, had less of an effect than the two longer isoforms 4 repeat plus 2 plus 3 and tau truncated at caspase cleavage site 421, 4 repeatC3 (+ p < 0.0001, post hoc Mann–Whitney tests 4 repeat minus 2 minus 3 vs. 4 repeat plus 2 plus 3 and 4 repeatC3), n = 35 cells. In the axons (b), expression of tau similarly decreases the area occupied by mitochondria, in the axon (#), and 4 repeat tau still has a more pronounced effect than 3 repeat tau (*). Within the 4 repeat isoforms, the 4 repeat tau construct truncated at caspase cleavage site 421 has a more pronounced effect than the other 4 repeat isoforms (+ p < 0.02, post hoc Mann–Whitney test 4 repeat C3 vs. 4 repeat plus 2 plus 3 or 4 repeat minus 2 minus 3). As in Fig. 2, the number of individual mitochondria was not quantified, only the amount of cytoplasm occupied by mitochondria in either the cell body (a) or the axon (b). The same cells were used in (a) and (b), first the axon was imaged, and followed back to cell body, which was then imaged. n = 35 cells per condition.

Constructs expressing five different isoforms of tau tagged with fluorescent proteins were generated with relative mobilities shown in a denaturing western blot (a). Constructs differed in the inclusion or exclusion of the splice changeable exons 2, 3 and 10 as well as one construct cleaved at amino acid 421 (4R + 2 + 3C3 construct) mimicking caspase 3 cleavage of tau (Gamblin et al. 2003). A schematic in (b) shows which exons are included in each construct.

Microfluidic devices with a design adapted from Taylor et al. (2005) were used to plate primary cortical neurons to isolate axons from cell bodies and dendrites [schematic shown in (a)]. (b) A photomicrograph of a microfluidic device with cultured neurons expressing CFP shows axons penetrating the microchannels. Scale bars 3 mm (a) 50 μm (b).

Tau over-expression increases the net movement of axonal mitochondria toward the neuronal cell body. (a) Shows a kymograph generated from a time series of mitochondria moving along an axon. Images were taken every 2 s and a total of 100 images were taken, for a total imaging session of 200 s. (b) Quantification of mitochondrial movements from the kymographs shows that velocity of mitochondria moving anterogradely is decreased (# p = 0.03, Mann-Whitney test control vs. tau), and that this reduction is more pronounced with 3 repeat tau than 4 repeat tau (* p < 0.0001 post hoc Mann–Whitney test). Average retrograde velocity (c) is not affected by tau over-expression. The percentage of mitochondria moving away from the cell body (d) is decreased with tau overexpression (# p < 0.0001 Mann-Whitney test). Conversely, the percentage of mitochondria moving retrogradely is increased with tau over-expression (# p < 0.0001, Mann-Whitney test), with 3 repeat tau having a more pronounced effect (* p < 0.0001 post hoc Mann–Whitney test) and with isoforms lacking N-terminal exons 2 and 3 having more of an effect than the other isoforms with the same number of repeats (+ p < 0.01 post hoc Mann–Whitney tests 3 repeat minus 2 minus 3 vs. 3 repeat plus 2 plus 3; 4 repeat minus 2 minus 3 vs. 4 repeat plus 2 plus 3; and 4 repeat minus 2 minus 3 vs. 4 repeat C3), n = 35 mitochondria from at least 10 individual neurons for each condition. The same kymographs were used to generate the data for (b), (c), and (d).