Figure 1.
Schematic representation of Tau and MAP2 expressed pan-neuronally in transgenic C. elegans.
(A) Diagrams of expressed proteins, full-length Tau (0N4R and 0N3R), Tau fragments (0N, 3R, and 4R), MAP2c, and MAP2c fragments (MAP2c amino-terminal and carboxyl-terminal) are shown. The Tau and MAP2c microtubule-binding domains (MTBDs) are depicted by blue and purple boxes, respectively. A 23-residue peptide in the UNC-119 amino-terminal assists proper expression and was used as the expression tag linked to the amino-terminus of Tau or MAP2. The numbers written above are the positions of the amino acids. (B) Comparison of the carboxyl-terminal amino acid sequences of Tau and MAP2. Identical amino acids are shown in red, and different amino acids are shown in black. Two perfect FTDP-17 mutations, Q336R and E342V, and two imperfect mutations, S325A (S325L in FTDP-17) and K369V (K369I in FTDP-17), that locate within the MAP2 MTBD are indicated.
Figure 2.
Neuronal dysfunction is induced by the carboxyl-terminal of Tau.
(A) Representative western blots showing the expressed full-length Tau (0N4R and 0N3R) and its fragments (0N, 3R, and 4R) in transgenic worms. Mock indicates the empty vector-transgenic worm line. Total lysates of the indicated 10 worms were subjected to western blotting using the UNC-119N antibody. The arrowhead indicates endogenous UNC-119. (B) The expression of each construct is presented as the relative abundance compared with the 0N4R/low-expressing worm (0N4R/low). n.s. indicates not significant. (C) Neurotoxicity was examined in each transgenic worm line. Data indicate the percentage of each worm line that showed uncoordinated movement (Unc). Note that the carboxyl-terminal fragment in 4R-Tg worms had a significantly higher percentage showing the Unc phenotype. 3R-Tg worms showed only a limited degree of Unc, whereas no Unc was observed in 0N-Tg worms. All data are presented as the mean±SEM from three independent experiments. Data were tested by one-way ANOVA followed by the Bonferroni–Dunn post hoc test. Asterisks indicate significance versus mock (P<0.005, significance level is 5%).
Figure 3.
Toxicity of both Tau and MAP2 in worm neurons.
(A) Representative western blot showing the expression of Tau and MAP2c in transgenic C. elegans, as probed with UNC-119N antibody. (B) The expression of each construct is presented as the relative abundance compared with the 0N4R/low-expressing line (0N4R/low). No significant difference was observed between the MAP2c and 0N4R/high lines. (C) Neuronal dysfunction is induced by MAP2c. Both MAP2c- and Tau 0N4R/high-expressing worms show a significantly higher percentage of worms with the Unc phenotype compared with the mock, indicating that MAP2c was as neurotoxic as Tau. (D) The carboxyl-terminal domain of MAP2 is responsible for its neurotoxicity. Unc analysis was performed using the MAP2c full-length (tmEx2557), MAP2c amino(N)-terminal (1 to 4 corresponded to tmEx3839,3840,3841 and 3842, respectively) and MAP2 carboxyl(C)-terminal (1 and 2 corresponded to tmEx3843 and 3844, respectively) Tg lines. The data are expressed as the mean±SEM from at least three independent experiments and were tested by one-way ANOVA followed by the Bonferroni–Dunn post hoc test. Asterisks indicate significance versus mock (P<0.005, significance level is 5%). # indicate significance versus MAP2c amino-terminal Tg lines (P<0.005, significance level is 5%).
Figure 4.
Biochemical characterizations of MAP2c and Tau expressed in transgenic C. elegans.
(A) Both MAP2c and Tau were highly phosphorylated in worm neurons. MAP2c and Tau (0N4R) were purified from the corresponding transgenic worms (MAP2c from tmIs849; 0N4R from tmIs390). Purified proteins were treated with or without phosphatase and subjected to western blotting using the HT7 (anti-human Tau monoclonal) and HM2 (anti-MAP2 monoclonal). (B) MAP2c and Tau did not bind to microtubules. The microtubules prepared were stabilized with taxol and GTP, and fractionated into the pellet (P) and supernatant (S). Both MAP2 and Tau remained in the supernatant (S). DM1A (anti-α-tubulin) and anti-UNC-119N (Tau and MAP2c) antibodies were used.
Figure 5.
Age-dependent neuritic abnormalities in Tau- or MAP2-expressing worms.
(A–E) CLSM images of neurites in the posterior part of the worm are shown. (A) Mock/DsRed-transgenic (Tg) worm, young. (B) Mock/DsRed-Tg worm, aged. (C) Tau(0N3R)/DsRed-Tg worm, young. (D) Tau(0N3R)/DsRed-Tg worm, aged. (E) Tau(0N4R)/DsRed-Tg worm, young. (F) Tau(0N4R)/DsRed-Tg worm, aged. (G) MAP2/DsRed-Tg worm, young. (H) MAP2/DsRed-Tg worm, aged. The scale bar is 100 µm. (I) Numbers of abnormal kinks (arrows) per 100 µm neurite. “Young” indicates 4–5 days after hatching, and “aged” indicates 10–11 days. The data are expressed as the mean±SEM. Asterisks indicate significant differences versus mock in each age group (one-way ANOVA followed by Bonferroni–Dunn post hoc test). # indicates a significant difference in the young versus aged group in the same line (P<0.05, Student's t-test). n = 21 to 23.
Figure 6.
MAP2 is not involved in the growth process of NFTs in the AD brain.
(A) Diagram of three site-specific anti-MAP2 antibodies and anti-Tau antibodies. (B) Human temporal cortex tissues from three AD patients and three normal controls were homogenized sequentially in detergent-containing buffers. Sarkosyl-insoluble, SDS-soluble fractions were prepared and subjected to SDS-PAGE followed by western blotting using anti-Tau antibodies (AT8 and PHF1) and three newly raised site-specific anti-MAP2 antibodies (MAP2-#39 and #40 are not shown). Total protein was used as the loading control. The staining intensity of Tau was increased markedly in the Sarkosyl-insoluble, SDS-soluble fractions from AD brains compared with normal brains. By contrast, the MAP2 antibodies failed to detect any increased patterns in AD brains compared with normal brains. Phosphatase treatment was performed to avoid effects from MAP2 phosphorylation. NC, normal brains; AD, Alzheimer's disease brains. Information about the cases is provided in Table S1. (C) Double immunofluorescence staining of the homologous carboxyl-terminal sequences of Tau and MAP2 in the AD brain. AD brain paraffin-embedded sections were double-labeled by anti-Tau antibody (PHF1) and anti-MAP2 antibody (MAP2-#41). Tau but not MAP2 localized in NFTs as well as in NTs. Representative Tau-positive-only neurons (arrowhead), MAP2-positive-only neurons (arrow) and Tau/MAP2-double-positive neurons (star) are indicated. Scale bars = 25 µm. (D) Average number of the three neuron types was counted per 640 µm2. The data are presented as the mean±SD.
Figure 7.
Tau but not MAP2c forms ThT-positive insoluble aggregates induced by heparin.
(A) The ThT fluorescence of Tau 0N4R isoform (circles), Tau 0N3R isoform (squares), and MAP2c (triangles) aggregates were measured at the times indicated. (B) After the 7-day incubation, the amount of Sarkosyl-insoluble proteins in the indicated samples was analyzed by SDS-PAGE followed by Coomassie brilliant blue staining.