Published as doi: 10.1096/fj.06-5721fje.
(The FASEB Journal. 2006;20:1272-1274.)
© 2006 FASEB
Tumor suppressor PTEN affects tau phosphorylation, aggregation, and binding to microtubules
Xue Zhang,
Feng Li,
Ayelen Bulloj1,
Yun-wu Zhang,
Gang Tong,
Zhuohua Zhang,
Francesca-Fang Liao and
Huaxi Xu2
Center for Neuroscience and Aging, Burnham Institute for Medical Research, La Jolla, California, USA
2Correspondence: Center for Neuroscience and Aging, Burnham Institute for Medical Research, La Jolla 92037, CA, USA. E-mail: xuh{at}burnham.org
SPECIFIC AIMS
The abnormally hyperphosphorylated tau is the major component of the neurofibrillary tangles (NFTs) that are found in a number of neurodegenerative diseases. In this study, we investigate the effects of tumor-suppressor chromosome 10 (PTEN) on tau phosphorylation, aggregation and binding to microtubules, and consequentially neuronal morphology.
PRINCIPAL FINDINGS
1. Overexpression of PTEN affects tau phosphorylation
In human tau and wild-type (WT) PTEN (PTEN–WT) cotransfected COS-7 cells, we examined 7 tau phosphorylation sites, including Akt targets Thr212 and Ser-214; GSK-3 targets Ser-199, Thr205, Thr212, and Ser-396; and Thr262 (Fig. 1
A). The levels of Thr212 and Ser-214 phospho-tau were significantly decreased by 
30 and 50%, respectively, compared with those in plasmid construct DNA control cells. Quantification of phospho-tau at other examined sites revealed increased phosphorylation of tau at Ser-199, Ser-202, and Thr205, and the phosphorylation of tau at Thr262 was virtually unchanged by overexpression of PTEN. We also examined the tau phosphorylation at these 7 sites in the cells coexpressing human tau and a phosphatase activity-null mutant PTEN, PTEN-CG. Compared with the control, the mutant PTEN selectively increased phospho-tau at Ser-396, a major tau phosphorylation site in PHF and NFTs, while leaving the other 6 sites unaltered (Fig. 1B, C
).
View larger version (28K):
[in this window]
[in a new window]
|
Figure 1. The effects of overexpression of PTEN proteins on tau phosphorylation and aggregation. A) Schematic drawing of human tau and major phosphorylation sites by various kinases. The sites examined are in bold. Kinases are abbreviated as following, M, MAPK; G, GSK-3; C5, CDK5; A, PKA; C, PKC; PK, phosphorylase kinase. B) pcDNA, wild-type (PTEN-WT) or lipid phosphatase null mutant PTEN (PTEN-CG) transfected COS-7 cells were analyzed by Western blots using phospho-tau specific antibodies as indicated. C) Levels of phospho-tau were quantified and normalized to total tau levels. Error bars indicate means ± SE, n=4, *P > 0.01 and **P > 0.02 compared with the respective controls (vector alone). D) COS-7 cells transiently overexpressing wild-type tau (T40) were further transfected with pcDNA vector, PTEN-WT, or PTEN-CG. Insoluble aggregated tau proteins were isolated and detected by using a filter-trap/immunoblotting assay (upper panels). Transfected PTEN and total tau proteins (as indicated in middle and lower panels) were detected by Western blots from same amounts of lysates as in filter assays. E) The amounts of aggregated tau protein were quantified by densitometry and normalized using the amounts of total tau. Error bars indicate means ± SE, n = 3, *P > 0.01 compared with the vector transfection control.
|
|
2. PTEN affects tau aggregation and microtubule-binding ability
The effect of PTEN on tau solubility was examined using a filter-retardation assay designed to detect protein (tau) aggregates. Insoluble tau molecules in cell lysates were retained in the filter after SDS washes, and the amount of insoluble tau was significantly decreased by 50% on overexpression of PTEN–WT as measured by immunoblotting of the filters. The aggregation of tau was increased 1.8-fold when the mutant PTEN was overexpressed (Fig. 1D, E
).
To test the effect of PTEN on tau’s partitioning into microtubules, we fractionated COS-7 cells cotransfected with human tau and either WT or the mutant PTEN to separate transfected tau proteins according to their intracellular localization. We found that PTEN-WT has little effect on the soluble cytosolic tau, while PTEN-CG reduced this population of tau by 30%. On the one hand, PTEN–WT increased the amount of microtubule-associated tau by 40%, concomitant with a 25% reduction in insoluble tau aggregation. On the other hand, expression of PTEN-CG resulted in a 40% increase in the insoluble membrane-bound tau and an 1.5-fold increase in aggregated tau.
3. The effects of PTEN on tau phosphorylation are mediated through PIP3 signaling pathway
We examined tau phosphorylation in a PTEN knock-out cell line that bears an inducible human Pten transgene with or without a PI3K inhibitor, LY294002 (LY). On induction of PTEN expression, tau phosphorylation at Ser214 was decreased by 40%, accompanied by reduced Akt activity. In the presence of LY, tau phosphorylation at Ser214 was further down-regulated (up to 80%) as well as Akt activity. In addition, Ser205 phospho-tau was increased by 20 or 40%, in the presence of induced PTEN or PI3K inhibitor, respectively.
In another experiment, we cotransfected COS-7 cells with WT PTEN and human tau, and treated the cells with insulin-like growth factor (IGF-1), LY or DMSO (control). We observed 30% more Ser214 phospho-tau in the presence of IGF-1, but an approximate 70% decrease in the LY- treated cells, compared with that in the DMSO control.
4. Phosphatase activity-null PTEN impairs neuronal tau binding to microtubules and alters neuronal morphology
We infected rat cortical primary neurons with Sindbis viruses expressing enhanced GFP (EGFP) alone, WT PTEN and EGFP, or the mutant PTEN and EGFP (Fig. 2
A). By staining endogenous tau and tubulin in the neurons, we found that tau and tubulin were completely overlapped in EGFP control or WT PTEN expressing neurons. However, in the neurons expressing the mutant PTEN, the majority of tau was concentrated in cell bodies and failed to colocalize with tubulin in dendrites (Fig. 2B
). In addition, the morphology of the neurons expressing the mutant PTEN was significantly altered with marked (60–70%) decreases in both the number and the length of dendrites (Fig. 2C
). Only slight increases in the number and the length of dendrites were observed when WT PTEN was expressed.
View larger version (39K):
[in this window]
[in a new window]
|
Figure 2. The lipid phosphatase null mutant PTEN impairs neuronal structure/neurite outgrowth. Rat cortical primary neurons were cultured for 2 wk and infected with sindbis viruses expressing GFP alone, GFP-IRES-wild-type PTEN or GFP-IRES-mutant PTEN. A) Expression of exogenous PTEN in infected primary neurons was confirmed by Western blots. Endogenous total tau and tubulin were also detected. B) Neurons expressing GFP alone (a–d), GFP-IRES-wild-type PTEN (e–h), or GFP-IRES-mutant PTEN (i–l) were visualized based on the GFP fluorescence (representing the expression of the transgenes), and further immunostained to detect tubulin (b, f, j) and tau (c, g, k). Fluorescence micrographs were visualized and recorded using a deconvolution microscope. d, h, and i are digitally merged images of tubulin and tau staining. C) Images of GFP-positive neurons were taken by fluorescence microscope (inset). The length and number of the neurons were counted. Results represent mean ± SE from at least 40 neurons of each experimental group. P > 0.05. Scale bars indicate 20 µm.
|
|
5. A decreased concentration of PTEN and increased tau phosphorylation at Ser214 in AD brains
We detected PTEN in 3 cases of AD patients by immunohistochemistry and found an 40% reduction of PTEN in the AD brains. The detection of Ser-214 epitope in AD but not the non-AD age-control brains was evident and found to be localized within NFTs in the fronto-forebrain neurons in all three cases of AD patient brains.
CONCLUSIONS AND SIGNIFICANCE
Collectively, we found tumor suppressor PTEN affects tau phosphorylation at multiple sites to regulate tau’s aggregation and binding to microtubules. For the first time we provide evidence for involvement of the PTEN-modulated PIP3 signaling pathway in the pathophysiological functions of tau. Our findings are in line with the previous studies showing that increased levels of active Akt are colocalized with NFTs in AD brain and in AD temporal cortex neurons, suggesting a molecular mechanism underlying tauopathy by defensive overactivation of the pro-survival Akt signaling through deregulation of PTEN.
PTEN is one of the dual specificity phosphatases (DSPs), possessing both protein and lipid phosphatase activity that are speculated as potential therapeutic targets for the treatment of both cancer and AD due to their roles in multiple signaling pathways. Our data support this new concept by showing that PTEN has different effects on different tau phosphorylation sites, which may be attributed to PTEN’s role in modulating varied tau kinases in different signaling pathways. Although we suggest that Ser214 phosphorylation may bea crucial major factor contributing to tau malfunction, the possible impacts of other PTEN-responsive phosphorylation sites on tau functions need further investigation. In addition, it is conceivable that PTEN may also regulate tau phosphorylation directly through its protein phosphatase activity. Therefore it remains to be further clarified whether tau is a direct substrate of PTEN.
In summary, our data demonstrate for the first time that tumor-suppressor PTEN can affect tau phosphorylation at different important PHF sites to regulate tau’s microtubule-binding function and aggregation. Mutations in Pten or deficiency in its phosphatase activity can lead to malfunction of tau. These findings delineate the link between the PIP3 pro-survival signaling pathway and tauopathy in neurodegeneration and potentially assign PTEN as a potential therapeutic target for AD.
View larger version (12K):
[in this window]
[in a new window]
|
Figure 3. PTEN regulates PIP3 signaling pathway to affect tau phosphorylation. The dashed lines indicate the putative mechanisms by which tau could be directly dephosphorylated by PTEN, and the feedback effects of cell death to provoke the Akt activity.
|
|
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.06-5721fje
1 Present address: Fundacion Instituto Leloir, Buenos Aires, Argentina 
Copyright © 2006 by The Federation of American Societies for Experimental Biology.
