22 March 2002 Unlike AβPP, whose physiological function remains nebulous, the protein tau is widely thought to promote neurite extension by stabilizing microtubules, the cytoskeletal struts that support outgrowing nerve processes. It is also credited with strengthening microtubule bundles running through the axons of adult neurons. Yet in the 18 March Journal of Cell Biology, researchers led by Eva-Maria Mandelkow of the Max-Planck-Unit for Structural Molecular Biology in Hamburg, Germany, extend that view by saying that only tiny amounts of tau are needed to serve these well-documented functions. Add more tau, and the neuron is in trouble.
The paper reports that elevated levels of tau inhibit the transport of vesicles and organelles in neurons by covering up the surface of microtubules and inhibiting the attachment of kinesin, a motor protein known to load cargo and ratchet its way along the microtubule toward the synapse. "This effect would set in simply by elevation of tau, long before pathological hyperphosphorylation or aggregation. Therefore, the process could speak to the issue of early defects in AD," write Eva-Maria and Eckhard Mandelkow in comments to ARF." In particular, it could explain why synapses decrease at a very early stage, before overt pathological signs of tau are visible. It also fits with the idea that defects in motors such as kinesin can have a similar outcome as elevated tau, because tau in the first instance inhibits kinesin-dependent processes. Thus, a function of tau that is normally considered 'good', namely microtubule binding and stabilization, can turn to be 'bad' at higher concentrations of tau because of transport inhibition," the Mandelkows add.
First author K. Stamer and colleagues carefully analyzed and quantified the effect of overexpressing tau in three different types of cultured neurons. In differentiated neuroblastoma cells, the number of peroxisomes-tiny organelles needed to detoxify reactive oxygen species such as hydrogen peroxide-found in neurites plummeted 17-fold when tau was overexpressed. Golgi-derived vesicles dropped fourfold, and mitochondria also nearly disappeared from neurites. This was true for hippocampal neurons cultured from rats or mice, as well.
No mitochondria means less ATP, and no peroxisomes means less catalase enzyme, two changes that left these neurites nearly defenseless against oxidative stress. In an experiment where control neuroblastoma cells retracted, and eventually lost, half their neurites upon exposure to hydrogen peroxide, neurite loss in tau-expressing neurons was 33-fold, down to almost none. Adding back the enzyme catalase partly rescued this loss. The study contains controls to make sure expression was not toxic by itself but indeed due to transport inhibition.
The amyloid precursor protein AβPP normally gets packaged into Golgi-derived vesicles and shipped down the axon to the nerve terminal, where some evidence suggests it exerts a neurotrophic function. Indeed, recent work from Larry Goldstein's lab suggests AβPP itself might dock these vesicles to kinesin (Kamal et al., 2000, and related ARF news item). Can tau, then, block AβPP transport? Indeed, neuroblastoma cells stably expressing human AβPP695, show AβPP-containing vesicles near the Golgi area and all along their neurite. Twelve hours after expressing tau in those neurons, tau appeared throughout the neurite while the number of AβPP vesicles in that same neurite had decrease threefold.
The authors also describe time-lapse video microscopy of cultured cortical neurons, in which they imaged tau and the transport of AβPP with two-color fluorescence in live neurons. AβPP vesicles was seen speeding down the axon at up to10 micrometers per second, more than 10 times faster than typical kinesin-driven traffic. Most fast transport occurred from the cell body toward the synapse, and 80 percent of all particles headed in this direction. Again, tau co-transfection stalled all that activity, essentially emptying axons of forward-moving vesicles and mitochondria within two days.
In earlier work, the Mandelkows had shown that the binding of tau to microtubules influences the rate with which kinesin attaches to and detaches from the microtubule (Trinczek et al, 1999).
The present study suggests a direct link between early events in tau and amyloid pathology, the authors write. When transport halts, AβPP vesicles will accumulate in the cell body near the trans-Golgi network, where Aβ peptide can be generated (Xu et al, 1997). "If the dwell time of AβPP were increased by a tau-dependent retardation of traffic, one would expect an increase in the production of Aβ with the known pathological consequences of aggregation and toxicity," the authors write.
Whether tau protein concentrations go up early on in the development of AD is unclear. The authors cite one study showing that tau protein levels are elevated in autopsy samples of Alzheimer's disease (Khatoon et al., 1992). They suggest the reason for this may lie in the neuron's tendency to sprout new neurites (for which tau would be required) in its attempt to protect itself from toxic challenges present in the aging brain (Savaskan & Nitsch, 2001).—Gabrielle Strobel
- Stamer K, Vogel R, Thies E, Mandelkow E, Mandelkow EM. Tau blocks traffic of organelles, neurofilaments, and APP vesicles in neurons and enhances oxidative stress. J Cell Biol. 2002 Mar 18;156(6):1051-63. PubMed.