The microtubule-associated protein tau is the major component of the intracellular neurofibrillary tangles (NFT) found in patients with various neurodegenerative disorders, including Alzheimer's and frontotemporal dementia. Scientists know that hyperphosphorylation of tau is at the root of its pathology, but what impact other proteins have on tau-related cell damage is less certain. In today's Neuron, George Jackson and colleagues at UCLA show that components of the wingless signal transduction pathway contribute to tau toxicity in flies.

Jackson et al. used a fruit fly model of tauopathy pioneered by Mel Feaney at Harvard Medical School (see ARF news item). In Jackson's hands, overexpression of human wild-type tau in the flies led to gross malformations of the compound eye. This effect was exacerbated by overexpression of glycogen synthase kinase-3 (GSK-3, the fly homolog is called shaggy), which is known to attenuate wingless signaling in Drosophila and has previously been fingered as a tau kinase and potential drug target in Alzheimer's (Anderton et al., 2000; Eldar-Finkelman, 2002; see ARF news item).

Interestingly, overexpression of another stepping stone in the wingless pathway, the fly homolog of human β-catenin, (armadillo), also exacerbated the effect of tau. This suggests that GSK-3's role in tauopathy lies outside of the wingless pathway, as one of its normal functions is to turn off armadillo in response to wingless signaling. Furthermore, the authors found that the transcription factor TCF, which is turned on by the wingless pathway, also exacerbated tau's damage to the eye.

Taken together, the data suggests that GSK-3 has dual, conflicting roles. On the one hand, it attenuates wingless signaling, and should, therefore, ameliorate the effects of armadillo and TCF on tau. But on the other hand shaggy itself exacerbates the effects of tau. The authors found that tau was hyperphosphorylated when co-expressed with GSK-3, making it tempting to conclude that the two directly interact.—Tom Fagan


  1. This paper describes an extremely important and exciting result that clearly demonstrates for the first time that both human tau as well as the Drosophila homolog of GSK3b, (a kinase implicated in abnormal tau phosphorylation) are essential for neurofibrillary pathology and neurodegeneration phenotype in Drosophila.

    In the current study, Jackson et al. overexpressed wild-type human 4-repeat tau. This resulted in neurodegeneration, as observed previously in Drosophila. However, when they expressed shaggy, a Drosophila homolog of GSK3b, the neurodegeneration was exacerbated. More importantly, it resulted in the formation of filamentous tau aggregates that are similar to the neurofibrillary tangles observed in Alzheimer's disease. Furthermore, the authors observed a correlation between the severity of neurodegeneration and the presence of hyperphosphorylated tau. In mice, Avila and colleagues have shown that inducible expression of GSK3b leads to tau hyperphosphorylation and neurodegeneration in hippocampus (Lucas et al., 2001).

    These results provide unequivocal evidence that abnormal tau phosphorylation per se is essential for initiating neurodegeneration. They also identify shaggy/GSK3b as an excellent candidate for mediating NFT formation in Drosophila. Modulation of downstream effectors of GSK3b in the wnt pathway, such as armadillo/b-catenin or pangolin/TCF, demonstrated that loss-of-function mutations suppressed while overexpression enhanced the tau-induced neurodegeneration. This supports the observation that GSK3b/shaggy is exerting its effects via direct tau phosphorylation rather than via the wnt pathway.

    This study provides critical support for therapeutic opportunities aimed at targeting proteins that lead to tauopathies. Obviously the next steps will include the use of inhibitors to reverse the observed phenotype."


    . Decreased nuclear beta-catenin, tau hyperphosphorylation and neurodegeneration in GSK-3beta conditional transgenic mice. EMBO J. 2001 Jan 15;20(1-2):27-39. PubMed.

  2. "Some excitement is in order following the Jackson et al study but commenter Rathan Bat should be reminded of some facts. Our experience in transgenic mice, and work by others in knockout mice, shows that extreme levels of GSK-3b are lethal, that is, both its absence and its overexpression (Spittaels et al., 2000; Spittaels et al., 1999; and work in press). You and I are living witness to the fact that intermediate levels are acceptable.

    It is not clear in the current Drosophila study what exactly the GSK-3b levels were in the flies? Are they "physiological" or rather, like many overexpression systems, highly un-physiological? Are they in any way similar to the levels in the brain of healthy humans or AD patients?

    GSK-3b was shown long ago to be a tau kinase in vitro, then we and others proved this in vivo in mammalian brain (Spittaels et al, 1999, 2000; Lucas et al., 2001). This is now once more confirmed in the fly. Not clear in any system is precisely how GSK-3b acts. Moreover, previous studies showed neurodegeneration in the fly caused by tau without any tangle formation. Also in mice, neurodegeneration due to AβPP does not require amyloid plaques per se.

    Taken together, all these data free us from the dogma that models must show plaques and tangles to be informative. Quite the opposite: I expect that models that do not show the classical AD lesions but "pre-pathological stages" will become more informative. On the other hand, the fact that the wingless pathway is not involved in the fly—or was it ?—only makes the problem harder to understand."


    . Glycogen synthase kinase-3beta phosphorylates protein tau and rescues the axonopathy in the central nervous system of human four-repeat tau transgenic mice. J Biol Chem. 2000 Dec 29;275(52):41340-9. PubMed.

    . Prominent axonopathy in the brain and spinal cord of transgenic mice overexpressing four-repeat human tau protein. Am J Pathol. 1999 Dec;155(6):2153-65. PubMed.

    . Decreased nuclear beta-catenin, tau hyperphosphorylation and neurodegeneration in GSK-3beta conditional transgenic mice. EMBO J. 2001 Jan 15;20(1-2):27-39. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. In Fly Model of Tauopathy, Neurons Degenerate Without Tangles
  2. Testosterone and Alzheimer's—a Possible Tau Connection

External Citations

  1. Anderton et al., 2000;
  2. Eldar-Finkelman, 2002

Further Reading

Primary Papers

  1. . Human wild-type tau interacts with wingless pathway components and produces neurofibrillary pathology in Drosophila. Neuron. 2002 May 16;34(4):509-19. PubMed.