For years, the two classic proteins of Alzheimer disease—Aβ and tau—seemed to move in separate orbits. Gradually, researchers are finding these proteins revolve more closely around each other than they had thought, as two new papers highlight. In a September 8 paper in the Journal of Neuroscience, researchers led by Eva-Maria and Eckhard Mandelkow at the Max-Planck Unit for Structural Molecular Biology, Hamburg, Germany, showed that when they added Aβ to healthy neuronal cultures, one of its earliest effects was to mislocalize tau into neuronal dendrites. In areas where tau was missorted, the authors also saw a dramatic loss of microtubules and synaptic spines, two effects believed to occur in early AD. Meanwhile, in the September 9 Science, researchers led by Lennart Mucke at the University of California, San Francisco, report that the loss of tau protects against axonal transport defects brought on by Aβ treatment. Both findings add intriguing new data to the nebulous picture of tau and Aβ interactions, although this area of investigation is still fraught with conflicting data and unanswered questions.

Previous work has shown that tau pathology occurs downstream of Aβ toxicity, and appears to be necessary for the latter, since the reduction of tau improves symptoms in AD model mice (see Rapoport et al., 2002 and ARF related news story on Roberson et al., 2007). However, another recent paper shows the opposite, finding that the loss of tau worsens symptoms in AD mice (see Dawson et al., 2010). These conflicting reports suggest that the interaction between the two proteins may be complicated. Research has begun to untangle the mechanisms behind the tau-Aβ relationship. For example, recent work led by Jürgen Götz at the University of Sydney, Australia, described a pathway by which dendritic tau mediates Aβ toxicity by helping supply synapses with a kinase that promotes excitotoxicity (see ARF related news story on Ittner et al., 2010). Most tau is axonal, however, and researchers led by George Bloom at the University of Virginia found that the soluble Aβ oligomers work via tau to destabilize microtubules, although the mechanism is unknown (see King et al., 2006). The Mandelkows’ group previously showed that when tau is not bound to microtubules, it is highly diffusible, and that overexpressing tau leads to its missorting into dendrites and the cell soma, and to the degeneration of synapses (see ARF related news story on Konzack et al., 2007 and Thies et al., 2007).

To try to pin down the mechanisms behind the Aβ/tau toxicity, Mandelkow and colleagues decided to examine the earliest changes in tau after application of soluble Aβ. First author Hans Zempel made hippocampal cultures from late-embryonic, healthy rats, then for three hours applied a solution of synthetic Aβ42 oligomers, made à la the ADDLs described by Bill Klein’s group (see Lambert et al., 1998). Zempel and colleagues found that Aβ treatment caused axonal tau to redistribute into the soma and dendrites in a small fraction (about 6 percent) of the neurons. In dendritic regions that gained tau, the authors saw several other changes that were not present in nearby dendrites lacking tau. Synaptic spine density and microtubule numbers fell by three-quarters or more in dendrites with tau, and the number of mitochondria dropped sharply as well. Neurofilaments in these dendrites increased, however, indicating they were missorted along with tau. The authors also saw increased calcium influx in these dendrites, and a greater degree of tau phosphorylation, mainly at phosphorylation sites that cause tau to detach from microtubules.

Since microtubule breakdown is a major feature of tau toxicity, the authors examined whether the addition of the microtubule-stabilizing agent taxol could reverse the changes they saw in dendrites containing tau. Zempel and colleagues found that taxol fully rescued the loss of microtubules, spines, and mitochondria, and also reduced the missorting of tau into dendrites. However, taxol had no effect on neurofilament missorting, nor on increased calcium influx or tau phosphorylation, suggesting that these effects are either upstream of microtubule loss, or are separately regulated.

Based on these data, the authors proposed a possible pathway for the early phases of Aβ-induced toxicity. They suggested that Aβ acts on N-methyl-D-aspartate (NMDA) receptors in the post-synapse to trigger calcium influx, which then leads to activation of kinases and phosphorylation of tau. This causes tau to let go of microtubules and drift away into dendrites. The authors speculated that calcium also acts to destabilize microtubules, and the loss of these neuronal highways stymies the normal trafficking of mitochondria and synaptic components, leading to the drop in synaptic spines. In support of the idea that calcium influx is one of the earliest events in this pathway, Zempel and colleagues found that several other cell stressors that are known to increase calcium influx, such as oxidative stress, excitotoxicity, and extracellular ATP, could produce the same effects as the application of Aβ.

“What I find most exciting is that [the authors] looked very carefully at individual cells and defined subregions of individual cells, rather than focusing on the cultures as a whole,” Bloom told ARF, pointing out that without careful observation, the authors could have missed these subtle early effects seen in only a handful of neurons. “[The paper] is revealing a new level of detail about what happens to neurons when they get exposed to certain kinds of β amyloid.”

The idea that calcium influx is an early event in Aβ-induced toxicity is a reasonable hypothesis, Bloom said. He speculated that tau itself may also play a role in signaling from Aβ. It would be very interesting to know whether the presence of tau is necessary for the downstream effects seen by Zempel and colleagues, he added. He suggested that an excellent follow-up experiment would be to perform the same studies on cells from tau knockout mice and compare the results.

In the second paper, Mucke and colleagues sought to resolve some of the conflicting findings regarding the effect of tau on axonal transport. Numerous studies have shown that Aβ impairs axonal transport (see Hiruma et al., 2003; Rui et al., 2006; ARF related news story on Pigino et al., 2009; and Decker et al.., 2010). But the picture for tau is much less clear. At the molecular level, researchers led by Erika Holzbaur at the University of Pennsylvania, Philadelphia, have shown that tau directly interferes with the motor proteins that walk down microtubules (see ARF related news story on Dixit et al., 2008). Work by the Mandelkow group has also shown that tau overexpression in neuroblastoma cells jams up microtubules and slows down intracellular transport of organelles such as mitochondria (see Ebneth et al., 1998). However, researchers led by Ralph Nixon at New York University showed that in mouse models, neither overexpression nor deletion of tau affected the rate of axonal transport in retinal neurons (see ARF related news story on Yuan et al., 2008).

Mucke and colleagues chose to examine the effect of tau in the presence of Aβ. First author Keith Vossel made hippocampal cultures from tau knockout mice, tau-deficient mice with one working copy of the tau gene, and wild-type mice. Vossel and colleagues added synthetic Aβ42 oligomers, made by reconstituting freeze-dried material, to the cultures and observed the rates of axonal transport of fluorescently labeled mitochondria and the growth factor receptor TrkA. Before Aβ treatment, the authors saw no differences in axon transport between any of the cultures. After Aβ treatment, the speed of axonal transport did not change, but the percentage of cargoes that were moving dropped by about half in the wild-type neurons. This effect was strongest for anterograde transport, and less pronounced for retrograde transport. In tau knockout and tau-deficient mice, however, Aβ42 oligomer treatment had no effect on cargo motility.

The results “provide another piece of evidence that tau reduction can make the brain more resistant against Aβ-induced deficits,” Mucke said. He suggested that the data might also help resolve some of the conflicting results in the field by showing that at baseline conditions, tau does not affect axonal transport, but that tau becomes important when cells are stressed with Aβ. Mucke said that they are following up on their finding by looking at the signaling machinery that regulates transport, to try to get at the mechanisms behind the beneficial effect of tau reduction. They will also look at other types of cargoes to see how widespread the effect on transport is.

The finding is significant, Holzbaur said, because it shows another link between Aβ and tau that suggests they share a common toxicity mechanism. One hypothesis in the field is that tau might affect the efficiency of APP processing and transport, Holzbaur said, so a logical next step might be to look at APP processing in tau knockouts versus wild-type mice. It would also be interesting to know if this Aβ-induced transport defect is enough to kill the neuron, or just to stress it, she said. She is not convinced, however, that the new paper resolves the debate about the role of tau on axonal transport. Holzbaur points out that the studies from her group and Nixon’s group were done on opposite ends of the scale—single molecule resolution versus the whole animal level, respectively. The data from Mucke and colleagues is not high resolution, Holzbaur said, and supports the finding from Nixon and colleagues that tau deletion or overexpression by itself does not change axonal transport. To resolve the discrepancy, she suggested, might require experiments at a middle range of resolution, perhaps by examining the transport of cargoes along microtubules.

Nixon does not believe the new findings resolve the issue of whether tau levels affect axonal transport, either, pointing out that the previous work looked at elevated tau levels as a pathogenic mechanism, while the data from Mucke and colleagues show that depleted tau levels might be a neuroprotective mechanism. “It’s apples and oranges,” Nixon said. He added that the paper is interesting because it extends the earlier finding by the Mucke group that the effects of Aβ are mediated by tau, by implicating axonal transport as part of the mechanism. Nixon would like to know whether this effect also occurs in vivo. He pointed out that tau’s effects are still mysterious, given that in some studies tau deletion is neuroprotective, and in other studies it is deleterious. “We know too little about what tau does,” Nixon said, “This line of investigation is crucial to understanding the early events [in AD].”—Madolyn Bowman Rogers


  1. This manuscript is very interesting. Mandelkow’s group basically has studied many aspects of tau, such as the structure, aggregation mechanism, phosphorylation kinases, and inhibitory effect of tau on axonal transport. But the mechanism of missorting of tau from axons to somatodendrites in degenerative neurons has remained an unsolved problem for a long time in the field of AD research. In this paper, they studied the effect of Aβ on tau translocation and concluded that toxic Aβ leads to tau missorting by inducing microtubule destabilization, because taxol treatment blocked this Aβ-induced tau missorting. If microtubule disassembly is the cause of tau missorting, then how does tau move from axon to somatodendrite after detaching from the microtubule? Traveling from axon to dendrite is such a long distance, it may be that this needs a transporter instead of occurring by simple Brownian motion.

    Interestingly, taxol prevents Aβ-induced missorting of tau and reduction of spine number without altering kinase activity. But MARK kinase phosphorylates the microtubule-binding region of tau, and induces dissociation of tau from microtubules. This raises interesting questions. Does Aβ activation of MARK not cause tau phosphorylation, or does phosphorylated tau stay on the microtubule?

    Synapse loss is the major cause of functional loss in the AD brain. If missorting of tau induces spine loss, there must be some underlying mechanisms. However, microtubule disassembly is required for tau missorting, which means that the pre-synapse may be lost through axonal degeneration when tau is missorted to the dendrites. Which occurs first, pre-synapse, or post-synapse loss?

    Most of papers start from the assumption that Aβ is a “bad guy,” and that increases in Aβ-induced synapse loss and neuron loss. Indeed, it is true in Aβ-treated cells and tissues, and in APP-Tg mice. However, removing Aβ cannot halt progression of AD. Aβ may have a physiological function, and when exaggerated, this function may cause AD dementia. However, we do not know whether Aβ has a physiological function or not. Even for tau, we do not know its physiological function other than microtubule stabilization. Studying the physiological function of key proteins in AD may help us to invent new therapeutic targets for the disease.

    View all comments by Akihiko Takashima
  2. While Aβ and tau exert separate modes of toxicity, it is fascinating to see how more and more details are revealed about how these toxic entities interact to impair neuronal functions in dementias.

    The new data presented by Lennart Mucke’s team in Science and by the Mandelkows in the Journal of Neuroscience address the fascinating interplay of Aβ and tau, the first study looking into axonal transport, and the second into sorting and morphological changes.

    Tau reduction has been shown by the Mucke team in 2007 to rescue, in vivo, from Aβ lethality (Roberson et al., 2007). This was followed by our study this July, identifying the kinase Fyn as a critical mediator in executing Aβ toxicity via tau (Ittner et al., 2010). Reducing Fyn in APP transgenic mice prevents Aβ toxicity, while overexpression enhances it (Chin et al., 2005; Chin et al., 2004).

    In the new study, Keith Vossel, Mucke, and colleagues transfected hippocampal neuronal cultures obtained from wild-type and tau-deficient mice with plasmids expressing fluorescent markers of mitochondria or the neurotrophin receptor TrkA to investigate axonal transport.

    Interestingly, upon incubating the cultures with oligomeric Aβ preparations, they found that tau reduction protected from Aβ-induced impairments in axonal transport. This extends the beneficial effects of a tau reduction reported previously (Ittner et al., 2010; Roberson et al., 2007) to axonal transport. What I find also interesting is that baseline anterograde transport seems to be slightly reduced in the heterozygous compared to the complete tau knockout, which may indicate that compensatory mechanisms kick in when tau is fully absent, but not when it is partially reduced. Aging could yet have another effect on any phenotype, as shown recently by crossing APP mutant mice onto a tau knockout background (Dawson et al., 2010)—loss of tau exacerbated pathology. Any pharmaceutical approach aimed at reducing rather than abolishing tau levels may have similar side effects.

    We previously found that both Aβ and tau cause impaired mitochondrial functions, both separately and synergistically (David et al., 2005; Rhein et al., 2009). While we looked at isolated cells or even isolated mitochondria, the advantage of the approach taken by Vossel et al. is that by imaging a tracking marker, the function of the neuron is maintained, which is different from when analyzing isolated cells or organelles as we have done. In a follow-up, it would be interesting to determine whether fibrillar preparations of Aβ cause the same effects as oligomers. For example, by looking at mitochondrial functions, such as complex activities, we previously did not detect dramatic differences between the two Aβ preparations (this was different from monomeric Aβ though, which had no effect on mitochondrial functions) (Eckert et al., 2008).

    Vossel and colleagues speculate that in addition to a tau reduction, components of the axonal transport machinery may be ideal targets. I fully agree with this notion, as we had previously found that JIP1, a component of the anterograde kinesin transport machinery, is trapped by phosphorylated tau in the cell, preventing the kinesin motors from transporting distinct cargoes to the axonal terminals (Ittner et al., 2008; Ittner et al., 2009). Interestingly, tau needs to be phosphorylated to cause this impairment, which is, incidentally, rescued by a small compound activating the tau phosphatase PP2A (van Eersel et al., 2010). This demonstrates that targeting phosphorylation of tau is a suitable strategy in treating tauopathies (Iqbal and Grundke-Iqbal, 2008).

    This leads me to the second paper, by Zempel and colleagues, who also used primary neuronal cultures to address the effects of Aβ oligomers on tau localization and phosphorylation. They found, upon incubating wild-type neurons with Aβ, that tau was sorted into the dendrite. We previously identified a crucial role for tau in the dendrite in executing Aβ toxicity via the NMDAR/PSD-95 complex (Ittner et al., 2010). The Mandelkow team also found that tau is differentially phosphorylated in axon and dendrite, a finding also made by us in transgenic mice overexpressing wild-type forms of human tau (Gotz and Nitsch, 2001). The work by Zempel and colleagues suggests lowering cytosolic calcium levels as an alternative strategy in treating AD.


    . Fyn kinase induces synaptic and cognitive impairments in a transgenic mouse model of Alzheimer's disease. J Neurosci. 2005 Oct 19;25(42):9694-703. PubMed.

    . Fyn kinase modulates synaptotoxicity, but not aberrant sprouting, in human amyloid precursor protein transgenic mice. J Neurosci. 2004 May 12;24(19):4692-7. PubMed.

    . Proteomic and functional analyses reveal a mitochondrial dysfunction in P301L tau transgenic mice. J Biol Chem. 2005 Jun 24;280(25):23802-14. PubMed.

    . Loss of tau elicits axonal degeneration in a mouse model of Alzheimer's disease. Neuroscience. 2010 Aug 11;169(1):516-31. Epub 2010 Apr 29 PubMed.

    . Oligomeric and fibrillar species of beta-amyloid (A beta 42) both impair mitochondrial function in P301L tau transgenic mice. J Mol Med (Berl). 2008 Nov;86(11):1255-67. PubMed.

    . Compartmentalized tau hyperphosphorylation and increased levels of kinases in transgenic mice. Neuroreport. 2001 Jul 3;12(9):2007-16. PubMed.

    . Alzheimer neurofibrillary degeneration: significance, etiopathogenesis, therapeutics and prevention. J Cell Mol Med. 2008 Jan-Feb;12(1):38-55. PubMed.

    . Parkinsonism and impaired axonal transport in a mouse model of frontotemporal dementia. Proc Natl Acad Sci U S A. 2008 Oct 14;105(41):15997-6002. Epub 2008 Oct 2 PubMed.

    . Dendritic function of tau mediates amyloid-beta toxicity in Alzheimer's disease mouse models. Cell. 2010 Aug 6;142(3):387-97. Epub 2010 Jul 22 PubMed.

    . Phosphorylated Tau interacts with c-Jun N-terminal kinase-interacting protein 1 (JIP1) in Alzheimer disease. J Biol Chem. 2009 Jul 31;284(31):20909-16. PubMed.

    . Amyloid-beta and tau synergistically impair the oxidative phosphorylation system in triple transgenic Alzheimer's disease mice. Proc Natl Acad Sci U S A. 2009 Nov 24;106(47):20057-62. PubMed.

    . Reducing endogenous tau ameliorates amyloid beta-induced deficits in an Alzheimer's disease mouse model. Science. 2007 May 4;316(5825):750-4. PubMed.

    . Sodium selenate mitigates tau pathology, neurodegeneration, and functional deficits in Alzheimer's disease models. Proc Natl Acad Sci U S A. 2010 Aug 3;107(31):13888-93. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. APP Mice: Losing Tau Solves Their Memory Problems
  2. Honolulu: The Missing Link? Tau Mediates Aβ Toxicity at Synapse
  3. Tau Roundup: Inducible Mice Accentuate Aggregation and More
  4. The Many Misdeeds of Aβ—Seizures and Axonal Transport Interference
  5. Axonal Transport Not Bothered by Tau Elevation In Vivo

Paper Citations

  1. . Tau is essential to beta -amyloid-induced neurotoxicity. Proc Natl Acad Sci U S A. 2002 Apr 30;99(9):6364-9. PubMed.
  2. . Reducing endogenous tau ameliorates amyloid beta-induced deficits in an Alzheimer's disease mouse model. Science. 2007 May 4;316(5825):750-4. PubMed.
  3. . Loss of tau elicits axonal degeneration in a mouse model of Alzheimer's disease. Neuroscience. 2010 Aug 11;169(1):516-31. Epub 2010 Apr 29 PubMed.
  4. . Dendritic function of tau mediates amyloid-beta toxicity in Alzheimer's disease mouse models. Cell. 2010 Aug 6;142(3):387-97. Epub 2010 Jul 22 PubMed.
  5. . Tau-dependent microtubule disassembly initiated by prefibrillar beta-amyloid. J Cell Biol. 2006 Nov 20;175(4):541-6. PubMed.
  6. . Swimming against the tide: mobility of the microtubule-associated protein tau in neurons. J Neurosci. 2007 Sep 12;27(37):9916-27. PubMed.
  7. . Missorting of tau in neurons causes degeneration of synapses that can be rescued by the kinase MARK2/Par-1. J Neurosci. 2007 Mar 14;27(11):2896-907. PubMed.
  8. . Diffusible, nonfibrillar ligands derived from Abeta1-42 are potent central nervous system neurotoxins. Proc Natl Acad Sci U S A. 1998 May 26;95(11):6448-53. PubMed.
  9. . Glutamate and amyloid beta-protein rapidly inhibit fast axonal transport in cultured rat hippocampal neurons by different mechanisms. J Neurosci. 2003 Oct 1;23(26):8967-77. PubMed.
  10. . Acute impairment of mitochondrial trafficking by beta-amyloid peptides in hippocampal neurons. J Neurosci. 2006 Oct 11;26(41):10480-7. PubMed.
  11. . Disruption of fast axonal transport is a pathogenic mechanism for intraneuronal amyloid beta. Proc Natl Acad Sci U S A. 2009 Apr 7;106(14):5907-12. PubMed.
  12. . Amyloid-beta peptide oligomers disrupt axonal transport through an NMDA receptor-dependent mechanism that is mediated by glycogen synthase kinase 3beta in primary cultured hippocampal neurons. J Neurosci. 2010 Jul 7;30(27):9166-71. PubMed.
  13. . Differential regulation of dynein and kinesin motor proteins by tau. Science. 2008 Feb 22;319(5866):1086-9. PubMed.
  14. . Overexpression of tau protein inhibits kinesin-dependent trafficking of vesicles, mitochondria, and endoplasmic reticulum: implications for Alzheimer's disease. J Cell Biol. 1998 Nov 2;143(3):777-94. PubMed.
  15. . Axonal transport rates in vivo are unaffected by tau deletion or overexpression in mice. J Neurosci. 2008 Feb 13;28(7):1682-7. PubMed.

Further Reading

Primary Papers

  1. . Abeta oligomers cause localized Ca(2+) elevation, missorting of endogenous Tau into dendrites, Tau phosphorylation, and destruction of microtubules and spines. J Neurosci. 2010 Sep 8;30(36):11938-50. PubMed.
  2. . Tau reduction prevents Abeta-induced defects in axonal transport. Science. 2010 Oct 8;330(6001):198. PubMed.