Global axonal transport rates are unaltered in htau mice in vivo

Abstract

Microtubule-based axonal transport is believed to become globally disrupted in Alzheimer's disease in part due to alterations of tau expression or phosphorylation. We previously showed that axonal transport rates along retinal ganglion axons are unaffected by deletion of normal mouse tau or by overexpression of wild-type human tau. Here, we report that htau mice expressing 3-fold higher levels of human tau in the absence of mouse tau also display normal fast and slow transport kinetics despite the presence of abnormally hyperphosphorylated tau in some neurons. In addition, markers of slow transport (neurofilament light subunit) and fast transport (snap25) exhibit normal distributions along optic axons of these mice. These studies demonstrate that human tau overexpression, even when associated with a limited degree of tau pathology, does not necessarily impair general axonal transport function in vivo.

Figure 1

Human tau over-expression and hyperphosphorylation in htau mice. A, B, Total optic nerve protein (20 μg) from mice after separation on 10% SDS gels were subject to immunoblot analysis with polyclonal anti-tau antibody LK recognizing only mouse tau (A) and polyclonal anti-tau antibody JM recognizing both human and mouse tau (B, C, D). Relative levels of tau isoforms on immunoblots determined by densitometry using NIH imaging were 3-fold higher in htau mice than in nontransgenic mice (B), consistent with elevations in brains from 8c mice. C, D, immunoblots of 100–200 μg of Triton-soluble and Triton-insoluble fractions from mouse retinas showing a partial shift in electrophoretic mobility of tau in htau mice, consistent with hyperphosphorylation. Levels of neurofilament protein (E) and snap25 (F) are unaltered. Immunocytochemistry with antibody PHF1 (Ps 396/404) detects accumulated hyperphosphorylated tau (arrows) in some retinal ganglion neurons of htau mice (H), but not in age-matched wild-type mice which exhibit no cytoplasmic staining (arrowheads) and pale non-specific nuclear staining (G). A fibrous inclusion (arrows) containing 15 nm straight filaments is seen in small numbers of retinal ganglion cells of htau mice (J and inset) but not in age-matched wild-type mice (I and inset). These lesions correspond in frequency to the incidence of very strong PHF1-positve neurons (J inset). Scale bars 7μm in G and H, 1μm in I and J, 3μm in J inset (top) and 200 nm in I and J insets (bottom).

Figure 2

Slow transport rates are not impaired in htau mice. At 3, 7 or 14 days after intravitreal injection of ³⁵S-methionine into wild-type (A, C) and htau mice (B, D), the optic pathways cut into 1-mm segments were fractionated into cytoskeleton (A, B) and soluble fractions (C, D), which were subject to 5–15% SDS-PAGE gels, transfer to nitrocellulose and autoradiography (A–D, at 14 days). Transport patterns were unaltered for either cytoskeletal or soluble proteins in htau mice. Redistribution of NF-M and tubulin along axons over time was quantified by densitometry scanning of autoradiographs. Graphs showing relative radioactivity of the labeled protein (vertical axis) plotted against distance from the eye displayed by optic segment number (horizontal axis) indicate that transport rates of labeled NF-M and tubulin are not altered in htau mice (E–J). Immunoblot analysis of the steady-state levels of NF-L along optic pathway reveals identical distribution in htau and wild-type mice (K, L, mean ± SD, n=4).

Figure 3

Fast transport rates in htau mice. Fast axonal transport was measured as described in Figure 2, except the post-injection time was 5 hours (A–D). Fast transport rates determined for protein p135, p85 and snap25 in Triton-soluble and insoluble fractions of optic axons were unaltered in htau mice (E–J). Fast transport marker snap25 determined by quantitative immunoblot analysis showed the same distribution in htau and wild-type mice (K, L). Wild-type and htau mice are represented by filled and unfilled circles, respectively.