Genetically augmenting tau levels does not ... - Wiley Online Library

6 downloads 137491 Views 1MB Size Report
in the 3xTg-AD mice has no effect on the onset and pro- gression of Ab .... using Graphpad Prism software (Graphpad Prism Inc., San Diego,. CA, USA). Results .... mice over-express the human tau gene six times higher relative to the ...
Journal of Neurochemistry, 2007, 102, 1053–1063

doi:10.1111/j.1471-4159.2007.04607.x

Genetically augmenting tau levels does not modulate the onset or progression of Ab pathology in transgenic mice Salvatore Oddo,* Antonella Caccamo,* David Cheng,* Bahareh Jouleh,  Reidun Torp  and Frank M. LaFerla* *Department of Neurobiology and Behavior and Institute for Brain Aging and Dementia, University of California, Irvine, California, USA  CMBN and Department of Anatomy, University of Oslo, Oslo, Norway

Abstract The two hallmark pathologies of Alzheimer’s disease (AD) are amyloid plaques, composed of the small amyloid-b (Ab) peptide, and neurofibrillary tangles, comprised aggregates of the microtubule binding protein, tau. The molecular linkage between these two lesions, however, remains unknown. Based on human and mouse studies, it is clear that the development of Ab pathology can trigger tau pathology, either directly or indirectly. However, it remains to be established if the interaction between Ab and tau is bidirectional and whether the modulation of tau will influence Ab pathology. To address this question, we used the 3xTg-AD

mouse model, which is characterized by the age-dependent buildup of both plaques and tangles. Here we show that genetically augmenting tau levels and hyperphosphorylation in the 3xTg-AD mice has no effect on the onset and progression of Ab pathology. These data suggest that the link between Ab and tau is predominantly if not exclusively unidirectional, which is consistent with the Ab cascade hypothesis and may explain why tauopathy-only disorders are devoid of any Ab pathology. Keywords: aging, Alzheimer’s disease, amyloid-b oligomers, neurofibrillary tangles, plaques. J. Neurochem. (2007) 102, 1053–1063.

Alzheimer’s disease (AD) is characterized by the accumulation of intraneuronal and extracellular amyloid-b (Ab) and by intracellular neurofibrillary tangles (NFTs), which are formed mainly by the hyperphosphorylated microtubule-binding protein tau (Selkoe 2001). In addition to AD, NFTs occur in other neurodegenerative disorders including frontotemporal dementia with Parkinsonism linked to chromosome 17, Pick’s disease, progressive supranuclear palsy, and corticobasal degeneration (Mandelkow and Mandelkow 1998; Higuchi et al. 2002). The mechanism by which tau exerts its neuronal toxicity is still controversial. Recently, it has been suggested that tau clogs axonal transport by interfering with motor proteins, prior to NFT formation (Mandelkow et al. 2003). This is consistent with evidence from other animal models showing that tau over-expression leads to neurodegeneration in absence of NFTs (Wittmann et al. 2001). Nevertheless, it is well established that tau hyperphosphorylation may alter microtubule stability and neuronal function (Mandelkow et al. 1995; Higuchi et al. 2002). The predominant hypothesis for the development of AD is the Ab cascade hypothesis, which stipulates that Ab is

the initiating cause of AD and other neuropathological changes occur as a consequence of Ab accumulation (Hardy and Selkoe 2002). This hypothesis is based on strong genetic evidence showing that mutations linked to early-onset AD increase Ab production (Citron et al. 1992; Suzuki et al. 1994). Recently, experiments performed in transgenic mice have supported the upstream role of Ab accumulation in the formation of NFTs (Gotz et al. 2001; Lewis et al. 2001; Oddo et al. 2004). Particularly, Gotz

Received February 6, 2007; revised manuscript received March 5, 2007; accepted March 7, 2007. Address correspondence and reprint requests to Frank M. LaFerla, Department of Neurobiology and Behavior, University of California, Irvine, 1109 Gillespie Neuroscience Bldg., Irvine, CA 92697-4545, USA. E-mail: [email protected] Abbreviations used: AD, Alzheimer’s disease; APP, amyloid precursor protein; Ab, amyloid-b; FA, formic acid; HSA, human serum albumin; NFTs, neurofibrillary tangles; PHF, paired helical filaments; PP2A, protein phosphatase 2A; SDS–PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis; TBNT, Trition X-100; TBS, Tris-buffered saline.

 2007 The Authors Journal Compilation  2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1053–1063

1053

1054 S. Oddo et al.

et al. (2001) have shown that injection of Ab into the brain of tau transgenic mice exacerbates tau pathology, whereas Lewis et al. (2001) have found enhanced tau pathology in mice expressing amyloid precursor protein (APP) and tau transgenes compared with single tau transgenic mice. We previously showed that clearance of Ab using Ab-specific antibodies reduced the tau burden (Oddo et al. 2004, 2006b,c). Taken together, these data strongly suggest a link between Ab and tau pathology and show that Ab accumulation can exacerbate tau pathology in vivo. It remains to be established, however, if the interaction between Ab and tau is bidirectional and whether the modulation of tau will influence Ab pathology. To address this issue, we used a transgenic model of AD (3xTg-AD) that develops both plaques and tangles in an age- and regional-dependent manner (Oddo et al. 2003b). We first determined the temporal relationship between tau phosphorylation and aggregation. Subsequently, we tested the consequences of genetically modulating tau levels on Ab pathology and found that increasing tau expression levels and phosphorylation has no effect on the onset and progression of Ab accumulation.

Material and methods

supernatant was stored as insoluble fraction. For sarkosyl extraction frozen tissue was prepared as previously described (Greenberg and Davies 1990). Immunohistochemistry Sections were stained as previously described (Oddo et al. 2006a). Briefly, 50 lm thick free-floating sections were used. The endogenous peroxidase activity was quenched for 30 min in 0.3% H2O2. When necessary, sections were then incubated in 90% FA for 7 min to expose the epitope. The appropriate primary antibody was applied overnight at 4C. Sections were subsequently washed in Trisbuffered saline (TBS) to remove excess of primary antibody and incubated with the appropriate secondary antibody for 1 h at 20C. After washing the excess of secondary antibody, sections were developed with diaminobenzidine substrate using the avidin–biotin horseradish peroxidase system (Vector Labs, Burlingame, CA, USA). Antibodies The following antibodies were used in this study: anti-Ab 6E10 (Signet Laboratories, Dedham, MA, USA), anti-Ab 1560 (Chemicon, Temecula, CA, USA), anti-Ab40 and anti Ab42 (Biosource, Camarillo, CA, USA), anti-Ab 35-40 (MM32-13.1.1, for Ab40) or anti-Ab 35-42 (MM40-21.3.4, for Ab42), anti-b-actin (Sigma, St Louis, MO, USA), anti-tau HT7, AT270, AT8, and AT100 (Pierce, Rockford, IL, USA), GSK3b-Ser9, protein phosphatase 2A (PP2A), and cdk5 (Biosource), anti-p25/p35 (Santa Cruz Biotechnology, Santa Cruz, CA, USA), b-Actin (Sigma-Aldrich), 12E8 was a generous gift from Dr Peter Seubert.

Mice The derivation of the 3xTg-AD and 2xTg mice used in this study has already been described (Oddo et al. 2003a,b). Briefly, the 3xTg-AD and the 2xTg mice were derived by co-microinjecting either two independent transgenes encoding human APPSwe and the human tauP301L (both under control of the mouse Thy1.2 regulatory element) or a single transgene encoding human tauP301L, respectively, into single-cell embryos harvested from homozygous mutant PS1M146V knockin (PS1-KI) mice. All animal procedures were in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals and in accordance with University of California, Irvine Institutional Animal Care and Use Committee recommendations. All appropriate measures were taken to minimize pain and discomfort in experimental animals.

Immunoblot For immunoblot, proteins from the soluble fraction were resolved by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS– PAGE) (10% Bis-Tris from Invitrogen) under reducing conditions and transferred to a nitrocellulose membrane. The membrane was incubated in a 5% solution of non-fat milk for 1 h at 20C. After overnight incubation at 4C with the primary antibody, the blots were washed in tween-TBS for 20 min and incubated at 20C with the secondary antibody. The blots were washed in T-TBS for 20 min and incubated for 5 min with Super Signal (Pierce). For the phosphatase treatment, proteins were pre-incubated for 1 h at 65C with phosphatase alkaline (Sigma).

Tissue preparation Mice were killed by CO2 asphyxiation and their brains were removed and cut sagittally. One hemibrain was fixed for 48 h in 4% paraformaldehyde and 50 lm thick free-floating sections were subsequently obtained using a vibratome slicing system (the vibratome company, St. Louis, MO, USA) and stored in phosphatebuffered saline and sodium azide. The other hemibrain was frozen in dry ice and subsequently homogenized in Tissue protein extraction reagent solution supplemented with complete Mini protease inhibitor tablet (Roche 1836153; Mannheim, Germany) and phosphatases inhibitors (Invitrogen, Carlsbad, CA, USA). The homogenized mixes were centrifuged at 4C for 1 h at 100 000 g. The supernatant was used for immunoblot analysis or as a soluble fraction of Ab ELISA measurements. The pellet was re-homogenized in 70% formic acid (FA) and centrifuged at 4C for 1 h at 100 000 g. The

Real time PCR Real time PCR was conducted as previously described (Kitazawa et al. 2005).

Enzymatic assays Protein phosphatase 2A activity was measured using a nonradioactive phosphatase assay system in accordance with the manufacturer’s instructions (Promega, Madison, WI, USA). GSK3b and cdk5 activities were measured as previously described (Kitazawa et al. 2005).

ELISA Enzyme-linked immunosorbent assay measurements were conducted as previously described (Oddo et al. 2005). Briefly, proteins from the soluble fraction (see above) were loaded directly onto ELISA plates and FA fractions were diluted 1 : 20 in neutralization

 2007 The Authors Journal Compilation  2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1053–1063

Augmenting tau levels does not modulate Ab pathology

buffer (1 mol/L Tris base; 0.5 mol/L NaH4PO4) prior to loading. MaxiSorp immunoplates (Nalge Nunc international, Rochester, NY, USA) were coated with monoclonal antibody 20.1, a specific antibody against Ab1–16 (from Van Nostrand, Stony Brook, NY, USA) in coating buffer (0.1 mol/L NaCO3 pH 9.6), and blocked with 3% bovine serum albumin. Synthetic Ab standards (Bachem, King of Prussia, PA, USA) were defibrillated by dissolving in hexafluoroisopropanol at 1 mg/mL and the hexafluoroisopropanol dissolving in hexafluoroisopropanol evaporated with a stream of N2. The defibrillated Ab was then dissolved in dimethylsulfoxide at 1 mg/mL. Standards of both Ab1–40 and Ab1–42 were made in antigen capture buffer (20 mmol/L NAH2PO4, 2 mmol/L EDTA, 0.4 mol/L NaCl, 0.5 g CHAPS, 1% bovine serum albumin, and pH 7.0), and loaded onto ELISA plates in duplicate. Samples were then loaded in duplicate and incubated overnight at 4C. Plates were washed and probed with either horseradish peroxidase-conjugated anti-Ab 35–40 (MM32-13.1.1, for Ab40) or anti-Ab 35–42 (MM4021.3.4, for Ab42) overnight at 4C. 3,3¢,5,5¢-tetramethylbenzidine was used as the chromagen, and the reaction was stopped with the addition of 30% O-phosphoric acid, and read at 450 nm on a plate reader (Labsystems, Sunnyvale, CA, USA).

saturated solution of NaOH in absolute ethanol for 2–3 s and incubated at 21C first for 10 min in 0.1% sodium borohydride and 50 mmol/L glycine in 5 mmol/L Tris buffer containing 0.3% NaCl and 0.1% Trition X-100 (TBNT), followed by incubation in 2% human serum albumin (HSA) in TBNT, and then for 12–16 h in mixtures of anti-paired helical filaments (PHF) in TBNT containing 2% HSA. Sections were then placed in 2% HSA in TBNT for 10 min followed by 2 h incubation in goat anti-rabbit Fab fragments coupled to 15 nmol/L gold particles (GFAR15; British BioCell International, Cardiff, UK) diluted 1 : 20 in TBNT containing 2% HSA and polyethylene glycol (0.5 mg/mL). Sections were examined and photographed in a Tecnai electron microscope. Statistical analysis The data were subsequently analyzed by ANOVA or t-test comparison, using Graphpad Prism software (Graphpad Prism Inc., San Diego, CA, USA).

Results

Age-dependent accumulation of NFTs To determine the relationship between tau phosphorylation and aggregation, we first stained sections from 6-, 12-, and 15-month-old 3xTg-AD mice with Gallyas silver stain. Gallyas-positive neurons are first readily apparent at 12 months of age and increase in an age-dependent fashion

Electron microscopy For post-embedding immunogold labeling, ultrathin Lowicryl HM20 section of freeze-substituted specimens obtained from perfusion fixed mouse brain were mounted on formvar-coated grids and processed for immunogold cytochemistry as previously described (Matsubara et al. 1996). Briefly, sections were immersed in a

(a)

1055

(b)

(d) (c)

Fig. 1 Age-dependent accumulation of neurofibrillary tangles (NFTs). (a–c) Representative sections from the hippocampus of 6-, 12-, and 15-month-old 3xTg-AD mice showing an age-dependent accumulation of NFTs, as detected by Gallyas silver staining. At 6 months of age, no Gallyas-positive neurons are apparent (a); it is not until the mice are 12 months of age that these lesions are readily detected (b). The number of NFTs detected by Gallyas progressively increases with age and is higher in 15-month-old mice (c). (d) Insol-

uble tau aggregates from the 3xTg-AD brains are also detected by biochemistry. Sarkosyl-insoluble tau is first detected at 12 months of age, paralleling the results obtained with the Gallyas staining. This representative blot was probed with AT100 and clearly shows a shift in the migration pattern between tau in the soluble (S) and insoluble (I) fraction; this shift is consistent with the slower mobility associated with a higher aggregation/phosphorylation state of the insoluble tau.

 2007 The Authors Journal Compilation  2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1053–1063

1056 S. Oddo et al.

(b)

(a)

Fig. 2 Ultrastructural characterization of neurofibrillary tangles (NFTs) in the 3xTg-AD mice. High and low magnification views of neurofibrillary tangles within pyramidal CA1 neurons of the hippocampus. (a) Arrowheads point to PHF-1 immunoreactive lesions within

the cytosol. (b) High magnification photograph showing the nature of the fibrils recognized with immunolabeling towards PHF-1. Scale bars, 500 and 100 nm for a and b, respectively. (N, nucleus, G, Golgi’s apparatus).

such that by 15 months of age, the majority of CA1 pyramidal neurons are Gallyas positive (Figs 1a–c). These data suggest that tau aggregation, as measured by Gallyas silver staining is first detected at 12 months of age. It has been shown that PHFs are also insoluble in different detergents, including Sarkosyl, whereas non-aggregated tau is soluble in this detergent. Consequently, this differential solubility in Sarkosyl is being used to differentiate between soluble and aggregated tau (Greenberg and Davies 1990). We next determined if the 3xTg-AD mice accumulate Sarkosylinsoluble tau aggregates. Using AT100 antibody, which detects phosphorylated tau at Thr212/Ser214, we first detect Sarkosyl-insoluble tau in 12-month-old 3xTg-AD mice (Fig. 1d). Notably, tau in the insoluble fraction runs slower

in a SDS–PAGE gel, consistent with biochemical alteration associated with phosphorylation and/or aggregation state (Fig. 1d). To further confirm the presence of NFTs in the 3xTg-AD mice, we performed immunoelectron microscopy and clearly identified NFTs accumulating into neurons of 3xTg-AD mice between 12 and 18 months of age (Figs 2a and b). In the 3xTg-AD mice, 12 months of age seems to be a critical point in the progression of tau pathology, as this corresponds to the first time when we can detect Gallyaspositive neurons and sarkosyl insoluble tau aggregates. We next evaluated Sarkosyl-soluble proteins extracted from 3xTg-AD mice at different ages by SDS–PAGE with or without prior alkaline phosphatase treatment. We found that tau extracted from 6-month-old mice was strongly positive

(a)

(c)

(e)

(b)

(d)

Fig. 3 Phosphorylation profile of soluble tau in the 3xTg-AD mice. To determine the phosphorylation profile of tau, we evaluated Sarkosyl-soluble proteins extracted from different aged 3xTg-AD mice by gel electrophoresis with or without prior alkaline phosphatase treatment. At 6 months of age, phosphorylation at the following sites is apparent: Thr212 and Ser214 as detected by antibody AT100 (a) and Ser262 as detected by the 12E8 antibody (b). At 12 months of age, phosphorylation at other sites in tau also becomes apparent: Thr181 as detected by the AT270 antibody (c), Ser202 and Thr205 as detected by the AT8 antibody (d), and Ser396/404 as detected by antibody PHF-1 (e). The arrow points to the 64 KDa-band detected by all the antitau antibodies used.

 2007 The Authors Journal Compilation  2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1053–1063

Augmenting tau levels does not modulate Ab pathology

by AT100 and to a less extent by 12E8. Notably the AT100 and 12E8 immunoreactivity was absent or markedly reduced in samples pre-treated with alkaline phosphatase (Figs 3a and b). These data indicate that phosphorylation at the AT100 and 12E8 sites occurs prior to tau aggregation and is first detected in 6-month-old mice, as we did not detect any tau phosphorylation in 2-month-old mice (data not shown). In contrast, we found that phosphorylation at the AT270, AT8, and PHF-1 sites are first apparent in 12-month-old mice, coinciding with the appearance of tau aggregation (Figs 3c–e). Tau in 6-month-old 3xTg-AD mice clearly seems more phosphorylated at the AT100 site compared with the 12E8 site. Although the functional significance of this differential phosphorylation is not yet clear, these data show that phosphorylation at Thr212/Ser214 (detected by the AT100 antibody) and phosphorylation at Ser262 (detected by the 12E8 antibody) occur before tau aggregation and represent an early event in the progression of the pathology in the 3xTg-AD mice. Age-dependent changes in cdk5 and GSK3b activity Several studies have implicated cdk5 and GSK3b as two of the major kinases involved in tau phosphorylation. To

Fig. 4 Age-dependent changes in GSK3b and cdk5 activity. (a) Representative blots showing age-dependent changes in the steadystate levels of GSK3b-Ser9, cdk5, p35, and p25. b-Actin was used as a loading control. (b) Densitometric analysis of the blots presented in (a), show an age-dependent decrease in the steady-state levels of the inactive form of GSK3b (n = 10/time point). The levels of GSK3b clearly decreased as a function of age, although the reduction was statistically significant only between 15 months versus the 6 months time-point (p < 0.001). (c) In contrast, the steady-state levels of cdk5 did not appear to change as a function of age in the 3xTg-AD mice (n = 10/time point). We next measured the levels of p35 and p25, two peptides that regulate cdk5 activity. Although the levels of p35 were

1057

determine how their activity changes as a function of age in the brains of the 3xTg-AD mice, we first measured the steady-state levels of GSK3b and cdk5. As previously reported, GSK3b activity is inhibited by the phosphorylation at Ser9 by AKT (Cross et al. 1995). Therefore, we measured the levels of GSK3b phosphorylated at Ser9 and found that the steady-state levels of the inactive form of GSK3b decreases as a function of age between 6 and 15 months of age (Figs 4a and b). These data suggest an age-dependent increase in GSK3b activity. We then measured the steady-state levels of cdk5 and found that the levels of this enzyme did not change as a function of age (Figs 4a and c). A small peptide p25, which is formed by the proteolytic cleavage of the cdk5-regulator p35, binds to cdk5 and causes its abnormal activation (Patrick et al. 1999). Although there is no broad consensus, p25 levels seems to be increased in AD brains (Patrick et al. 1999; Nguyen et al. 2002). To determine if p35 and p25 levels change as a function of age in the brains of the 3xTg-AD mice, we measured the levels of these peptides by western blot. Although p35 levels are similar across the ages tested (Figs 4a and d), we found that the levels of p25 change as a function of age, with an age-dependent increase from

unchanged throughout all the time points analyzed (d), we found that p25 levels increased as a function of age reaching peak levels at 15month old (n = 10/time point) (e). p25 levels were significantly higher between 6 and 15 months of age (p < 0.05). To directly determine how the activity of GSK3b and cdk5 change as a function of age, we measured their activity using specific substrates in the presence of 32 P. We found that GSK3b activity was significantly higher in 15month-old 3xTg-AD mice compared with age-matched non-Tg mice (p < 0.05) (n = 10/time point) (f). Similarly, cdk5 activity was higher in 15-month-old 3xTg-AD mice compared with age-matched non-Tg mice (p < 0.05) (n = 10/time point) (g).

 2007 The Authors Journal Compilation  2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1053–1063

1058 S. Oddo et al.

Fig. 5 Protein phosphatase 2A (PP2A) activity is not altered in the brains of 3xTg-AD mice. To measure the activity of PP2A as a function of age, we used a non-radioactive phosphatase assay system. We found that the activity of this enzyme did not change as a function of age in the 3xTg-AD or non-Tg mice (n = 10/time point).

6 to 15 months of age (Figs 4a and e). These data suggest that cdk5 activity increases as a function of age. To directly measure the enzymatic activity of GSK3b and cdk5, we incubated protein extracts from the 3xTg-AD and Non-Tg mice with specific substrates of these kinases in the presence of 32P. The amount of the incorporated 32P reflects the kinase activity. As for the steady-state levels of GSK3b-Ser9, the activity of this enzyme peaks at 15 months of age (Fig. 4f). Cdk5 activity also changes as a function of age, and it is significantly higher in the 3xTgAD compared with the Non-Tg at 15 months of age (Fig. 4g). Notably, cdk5 activity was higher in 15-monthold versus 12-month-old 3xTg-AD mice, which is consistent with the p25 levels, also higher in 15-month-old mice (Figs 4e and g). Taken together, these data indicate that altered GSK3b and cdk5 activity can contribute to the tau pathology in the 3xTg-AD mice, although the initial tau phosphorylation seems to be independent of altered GSK3b and cdk5 activity. Tau hyperphosphorylation can occur either because of an increase in the activity of one or more kinases or because of decreased phosphatase activity. A major phosphatase implicated in tau dephosphorylation is PP2A (Trojanowski and Lee 1995; Liu et al. 2005; Tanimukai et al. 2005). To determine if this enzyme changes as a function of age, we directly measured PP2A activity and found that its activity was similar between 3xTg-AD and Non-Tg mice between 1 and 15 months of age (Fig. 5). These results indicate that at the ages analyzed, PP2A activity is not altered in the 3xTgAD mice and therefore is not likely to contribute to tau pathology in these mice. Increased tau levels and phosphorylation in the 3xTg-ADh/tau mice We next sought to determine if the interaction between Ab and tau is bidirectional and whether modulation of tau steady-state levels will influence Ab pathology. To increase

Fig. 6 Schematic representation of the breeding strategy. To genetically increase tau expression levels, we crossed the 3xTg-AD mice with double transgenic mice over-expressing the human tauP301L gene and harboring the M146V mutation in the PS1 gene. Fifty percent of the offspring are heterozygous for the PS1M146V mutation and hemizygous for the human amyloid precursor protein (APP) and tau genes. In contrast, the other 50% of the offspring will have an extra copy of tau deriving from the double transgenic mice. Notably, as both the 3xTg-AD and 2xTg mice are on the same genetic background, the 3xTg-ADh/tau mice and the 3xTg-ADh mice derived from these crosses are on the same genetic background.

tau pathology and to determine its effects on the onset of the Ab pathology, we crossed homozygous 3xTg-AD mice with double transgenic mice harboring the human PS1M146VKI mutation and the tauP301L transgene (PS1/tau). The PS1/tau mice over-express the human tau gene six times higher relative to the endogenous mouse tau gene, which is comparable with tau levels in homozygous 3xTg-AD mice (Oddo et al. 2003b). All the mice obtained from these crosses were homozygous for the PS1M146V mutation and hemizygous for the human APP gene (Fig. 6). In addition 50% of the mice had two tau transgenes, one inherited from the 3xTg-AD mice and one from the 2xTg mice (these mice are indicated as ‘3xTg-ADh/tau’). As expected, the 3xTg-ADh/tau mice showed increased steady-state levels of human tau compared with age- and sex-matched 3xTg-AD hemizygous (3xTg-ADh) littermates both at 6, 12, and 15 months of age (Figs 7a and b). In addition, we also found that the increase in the steady-state levels of tau leads to an increase in tau phosphorylation at Thr181 as detected by the anti-tau antibody AT270 (Figs 7a and c). Notably, we also found that phosphorylation at Thr231 was only significant between 15-month-old 3xTg-ADh and 3xTg-ADh/tau, as detected by anti-tau antibody AT180 (Figs 7a and d). These data indicate that in the 3xTg-ADh/tau phosphorylation of tau occurs in a hierarchical manner, with phosphorylation at Thr181 occurring earlier than phosphorylation at Thr231. The increase in tau levels and phosphorylation were confirmed by immunohistochemistry. We found that the somatodendritic levels of total tau (detected by the HT7

 2007 The Authors Journal Compilation  2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1053–1063

Augmenting tau levels does not modulate Ab pathology

1059

Fig. 7 Higher tau levels and phosphorylation in the 3xTg-ADh/tau mice. (a) Representative western blots in 6- and 12-month-old mice. (b) Densitometric analysis of the western blots show that the steadystate levels of tau, as detected by antibody HT7, were significantly higher in the 3xTg-ADh/tau mice compared with age- and gendermatched 3xTg-ADh mice at all ages analyzed. (c). Densitometric analysis of the blots probed with antibody AT270, which recognize

phosphorylated tau at Thr181, showed an increase in tau phosphorylation in the 3xTg-ADh/tau mice. Although this difference was only significant between 12- and 15-month-old mice, a clear but not significant trend was evident for 6-month-old mice. (d) In contrast, phosphorylation of tau at Thr231, as detected by antibody AT180, was only changed in 15-month-old mice (n = 10/time point).

antibody) and phosphorylated tau (detected by the AT270 antibody) were increased in the hippocampus of 6-, 12-, and 15-month-old 3xTg-ADh/tau compared with age-matched 3xTg-ADh littermates (Figs 8a–l). In contrast tau phosphorylated at Thr231 was different between these two groups only in 15-month-old mice (Figs 8m–r). These data indicate that increasing tau expression levels leads to an increase in tau phosphorylation. To determine if the increase in tau phosphorylation in the 3xTg-ADh was mediated by GSK3b and cdk5, we measured the levels of these enzymes in the brains of the 3xTg-ADh/ tau and their age- and gender-matched 3xTg-ADh littermates. We found that the levels of GSK3b-Ser9 and cdk5 were similar between the two groups at all the time points analyzed (Figs 9a–c). In contrast, we found that the steadystate levels of p35/p25 were significantly different between these two groups. In particular, we found that p35 levels decreased as a function of age in the 3xTg-ADh/tau mice whereas, they were unaltered in the 3xTg-ADh mice (Figs 9a

and d). The p35 levels where statistically significant at 12 and 15 months of age between the two groups. Similarly, p25 levels were significantly higher in 12- and 15-month-old 3xTg-ADh/tau compared with 3xTg-ADh mice (Figs 9a and e). These data suggest that the increase in tau phosphorylation in the 3xTg-ADh/tau mice may be mediated by an increase in cdk5 activity. Increasing tau levels and phosphorylation does not alter Ab pathology We next determined if the increase in tau pathology in the 3xTg-ADh/tau mice leads to any changes in APP processing and/or Ab deposition. We first measured the steady-state levels of APP and its C-terminal fragments C99 and C83 by western blot and found that these levels were similar between the 3xTg-ADh and the 3xTg-ADh/tau mice (Figs 10a–d). We then directly measured Ab levels using sandwich ELISA. Insoluble Ab levels were below detection levels in 6- and 12-month-old mice (data not shown). This is consistent with

 2007 The Authors Journal Compilation  2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1053–1063

1060 S. Oddo et al.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

(j)

(k)

(l)

(m)

(n)

(o)

(p)

(q)

(r)

Fig. 8 Enhanced somatodendritic accumulation of tau in the 3xTgADh/tau mice. Representative microphotographs showing the CA1 region of 3xTg-ADh and 3xTg-ADh/tau mice stained with antibodies HT7, AT270, and AT180. (a–f) Somatodendritic accumulation of tau was clearly increased in the 3xTg-ADh/tau mice compared with ageand gender-matched 3xTg-ADh mice. (g–l) Similarly, the levels of phosphorylated tau at Thr181 in the somatodendritic compartment of

CA1 pyramidal neurons were higher in the 3xTg-ADh/tau mice compared with the 3xTg-ADh mice. (m–r) In contrast, AT180 immunoreactivity was similar between the two groups at 6 and 12 months of age. Fifteen-month-old 3xTg-ADh/tau mice, however, showed a marked increase in AT180 immunoreactivity compared with age- and gendermatched 3xTg-ADh mice.

the fact that the mice in both groups are hemizygous for the human APPSwe transgene. In contrast, we found that both soluble Ab40 and Ab42 levels were not significantly changed between the two groups (Figs 11a and b). Similarly, insoluble Ab levels were not significantly different between 15-month-old 3xTg-AD/h and 3xTg-ADh/tau mice (data not shown). To determine if the increase in tau pathology modulates the onset of the Ab deposition, we stained sections from the 3xTg-ADh and 3xTg-ADh/tau mice with the monoclonal antibody 1560, which is raised against the first 17 amino acids of the Ab sequence and found that the number of Ab deposits were similar between the two groups of mice at both ages analyzed (Figs 11c–h). These results are consistent with recent in vitro experiments showing that tau transfection does not increase Ab production in neuronal cell lines (Goldsbury et al. 2006). Taken together, these data indicate that the increase in tau levels and phosphorylation has no effect on the Ab levels, strongly suggesting that the interaction between Ab and tau pathology is unidirectional.

studies showing that tau phosphorylation at residues 12E8 and AT100 occurs early in the disease progression whereas phosphorylation at other sites, such as PHF-1 seems to occur at later stages. Notably, the 12E8 and PHF-1 immunoreactivity was reduced in 15-month old compared with 12-month old 3xTg-AD mice. Although further studies are necessary to better understand the reason behind such decrease in immunoreactivity, we are tempted to speculate that a conformational change in the structure of these tau deposits may occur as the mice age, thus masking the epitopes for the 12E8 and PHF-1 antibodies. The data presented here show that increasing tau expression and phosphorylation in a transgenic model has no effects on the onset and progression of the Ab pathology. This is consistent with the human genetic data. In this regard, mutations in the APP, PS1, and PS2 genes lead to early-onset AD by altering APP processing and increasing Ab production, strongly suggesting that Ab accumulation is upstream in the cascade of events leading to AD, including the accumulation of NFTs. In contrast, mutations in the tau gene lead to neurodegeneration in the absence of Ab accumulation, suggesting that an increase in tau accumulation does not lead to an increase in Ab levels. The amyloid cascade hypothesis states that Ab accumulation is the initial trigger of AD and that other neuropathological changes occurs as a consequence of Ab pathology. Several studies have corroborated this hypothesis. In this regard, transgenic animal models have been very helpful and have added strong data supporting the amyloid cascade hypothesis. In particular, we have previously shown that Ab immunotherapy can reverse the

Discussion

In these studies, we demonstrate that tau pathology develops in an age-dependent fashion and there is a hierarchical phosphorylation of tau at different residues, in the 3xTg-AD mice. We showed that phosphorylation at residues AT100 and 12E8 precedes tau aggregation, whereas phosphorylation at sites recognized by AT270, PHF-1, and AT8 coincides with the detection of aggregated tau as showed by Gallyas silver staining. This finding is consistent with previous

 2007 The Authors Journal Compilation  2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1053–1063

Augmenting tau levels does not modulate Ab pathology

1061

Fig. 9 cdk5 mediates the increase in tau phosphorylation in the 3xTg-ADh/tau mice. (a) Representative blots showing the steady-state levels of GSK3b-Ser9, cdk5, p35, and p25. b-Actin was used as a loading control. (b and c) Densitometric analysis of the blots presented in (a), show that the steady-state levels of GSK3b-Ser9 and cdk5 were

similar between the 3xTg-ADh and the 3xTg-ADh/tau mice. (d) In contrast, the levels of p35 were significantly higher in 12- and 15month-old 3xTg-ADh mice compared with 3xTg-ADh/tau mice. (e) The changes in p35 levels were reflected by an increase in p25 levels in the 3xTg-ADh/tau mice (n = 10/time point).

early tau pathology and rescue the behavioral deficits in a transgenic model with both plaques and tangles (Oddo et al. 2004, 2006b), strongly suggesting that Ab accumulation facilitates the formation of neurofibrillary pathology in the 3xTg-AD mice. Similar results have been obtained with other transgenic animals. In particular it has been shown that NFTs can be induced by injecting tau transgenic mice with Ab42 (Gotz et al. 2001). Similar

conclusions have been reached by breeding APP transgenic mice to tau transgenic mice (Lewis et al. 2001; Ribe et al. 2005). These studies, together with the results presented in this manuscript, strongly suggest that the link between Ab and tau is unidirectional, with modulation of Ab having a direct effect on tau pathology whereas modulation of tau does not seem to alter Ab deposition in this experimental paradigm.

 2007 The Authors Journal Compilation  2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1053–1063

1062 S. Oddo et al.

(a)

(b)

(c)

(d)

Fig. 10 The increase in tau pathology did not change amyloid precursor protein (APP) processing. (a) Representative western blots probed with antibody 6E10, to detect full length APP, with antibody CT20 to detect C99 and C83 levels, and with antibody b-Actin used as

(a)

a loading control. (b–d) Densitometric analysis of the western blots show that the steady-state levels of APP, C99, and C83 were similar between the 3xTg-ADh/tau and 3xTg-ADh mice, at all ages analyzed (n = 10/time point).

(b)

(c)

(d)

(e)

(f)

(g)

(h)

Fig. 11 The increase in tau pathology did not change amyloid-b (Ab) levels and deposition. To determine if the steady-state levels of soluble Ab40 and Ab42 were altered between the 3xTg-ADh/tau and the 3xTg-ADh mice, we performed sandwich ELISA. (a,b) The levels of soluble Ab40 and Ab42 were not statistically significant between the

two groups of mice. (c–h) Representative microphotograph showing hippocampal sections from 3xTg-ADh and 3xTg-ADh/tau mice. The Ab load was similar between the two groups of mice at all the time points analyzed (n = 10/time point).

 2007 The Authors Journal Compilation  2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1053–1063

Augmenting tau levels does not modulate Ab pathology

Acknowledgements We thank Mrs Levina Tran, Mrs Lana Tran for excellent technical support and Dr Mathew Blurton-Jones for helpful discussion. We thank Dr Peter Seubert at Elan Corporation for kindly providing the 12E8 antibody. This work was supported by funding from the NIA (AG0212982) to F.M.L. and funding from Functional Genomics in Norway, (FUGE) to R.T.

References Citron M., Oltersdorf T., Haass C., McConlogue L., Hung A. Y., Seubert P., Vigo-Pelfrey C., Lieberburg I. and Selkoe D. J. (1992) Mutation of the beta-amyloid precursor protein in familial Alzheimer’s disease increases beta-protein production. Nature 360, 672–674. Cross D. A., Alessi D. R., Cohen P., Andjelkovich M. and Hemmings B. A. (1995) Inhibition of glycogen synthase kinase-3 by insulin mediated by protein kinase B. Nature 378, 785–789. Goldsbury C., Mocanu M. M., Thies E., Kaether C., Haass C., Keller P., Biernat J., Mandelkow E. and Mandelkow E. M. (2006) Inhibition of APP trafficking by tau protein does not increase the generation of amyloid-beta peptides. Traffic 7, 873–888. Gotz J., Chen F., van Dorpe J. and Nitsch R. M. (2001) Formation of neurofibrillary tangles in P301l tau transgenic mice induced by Abeta 42 fibrils. Science 293, 1491–1495. Greenberg S. G. and Davies P. (1990) A preparation of Alzheimer paired helical filaments that displays distinct tau proteins by polyacrylamide gel electrophoresis. Proc. Natl Acad. Sci. USA 87, 5827– 5831. Hardy J. and Selkoe D. J. (2002) The amyloid hypothesis of Alzheimer’s disease: progress and problems on the road to therapeutics. Science 297, 353–356. Higuchi M., Lee V. M. and Trojanowski J. Q. (2002) Tau and axonopathy in neurodegenerative disorders. Neuromolecular Med. 2, 131– 150. Kitazawa M., Oddo S., Yamasaki T. R., Green K. N. and LaFerla F. M. (2005) Lipopolysaccharide-induced inflammation exacerbates tau pathology by a cyclin-dependent kinase 5-mediated pathway in a transgenic model of Alzheimer’s disease. J. Neurosci. 25, 8843– 8853. Lewis J., Dickson D. W., Lin W. L. et al. (2001) Enhanced neurofibrillary degeneration in transgenic mice expressing mutant tau and APP. Science 293, 1487–1491. Liu F., Grundke-Iqbal I., Iqbal K. and Gong C. X. (2005) Contributions of protein phosphatases PP1, PP2A, PP2B and PP5 to the regulation of tau phosphorylation. Eur. J. Neurosci. 22, 1942–1950. Mandelkow E. M. and Mandelkow E. (1998) Tau in Alzheimer’s disease. Trends Cell Biol. 8, 425–427. Mandelkow E., Song Y. H., Schweers O., Marx A. and Mandelkow E. M. (1995) On the structure of microtubules, tau, and paired helical filaments. Neurobiol. Aging 16, 347–354. Mandelkow E. M., Stamer K., Vogel R., Thies E. and Mandelkow E. (2003) Clogging of axons by tau, inhibition of axonal traffic and starvation of synapses. Neurobiol. Aging 24, 1079–1085. Matsubara A., Laake J. H., Davanger S., Usami S. and Ottersen O. P. (1996) Organization of AMPA receptor subunits at a glutamate

1063

synapse: a quantitative immunogold analysis of hair cell synapses in the rat organ of Corti. J. Neurosci. 16, 4457–4467. Nguyen K. C., Rosales J. L., Barboza M. and Lee K. Y. (2002) Controversies over p25 in Alzheimer’s disease. J. Alzheimers Dis. 4, 123–126. Oddo S., Caccamo A., Kitazawa M., Tseng B. P. and LaFerla F. M. (2003a) Amyloid deposition precedes tangle formation in a triple transgenic model of Alzheimer’s disease. Neurobiol. Aging 24, 1063–1070. Oddo S., Caccamo A., Shepherd J. D., Murphy M. P., Golde T. E., Kayed R., Metherate R., Mattson M. P., Akbari Y. and LaFerla F. M. (2003b) Triple-transgenic model of Alzheimer’s disease with plaques and tangles: intracellular Abeta and synaptic dysfunction. Neuron 39, 409–421. Oddo S., Billings L., Kesslak J. P., Cribbs D. H. and LaFerla F. M. (2004) Abeta immunotherapy leads to clearance of early, but not late, hyperphosphorylated tau aggregates via the proteasome. Neuron 43, 321–332. Oddo S., Caccamo A., Green K. N., Liang K., Tran L., Chen Y., Leslie F. M. and LaFerla F. M. (2005) Chronic nicotine administration exacerbates tau pathology in a transgenic model of Alzheimer’s disease. Proc. Natl Acad. Sci. USA 102, 3046–3051. Oddo S., Caccamo A., Smith I. F., Green K. N. and Laferla F. M. (2006a) A dynamic relationship between intracellular and extracellular pools of Ab. Am. J. Pathol. 168, 184–194. Oddo S., Vasilevko V., Caccamo A., Kitazawa M., Cribbs D. H. and Laferla F. M. (2006b) Reduction of soluble Abeta and tau, but not soluble Abeta alone, ameliorates cognitive decline in transgenic mice with plaques and tangles. J. Biol. Chem. 281, 39413– 39423. Oddo S., Caccamo A., Tran L., Lambert M. P., Glabe C. G., Klein W. L. and LaFerla F. M. (2006c) Temporal profile of amyloid-beta (Abeta) oligomerization in an in vivo model of Alzheimer disease. A link between Abeta and tau pathology. J. Biol. Chem. 281, 1599–1604. Patrick G. N., Zukerberg L., Nikolic M., de la Monte S., Dikkes P. and Tsai L. H. (1999) Conversion of p35 to p25 deregulates Cdk5 activity and promotes neurodegeneration. Nature 402, 615–622. Ribe E. M., Perez M., Puig B. et al. (2005) Accelerated amyloid deposition, neurofibrillary degeneration and neuronal loss in double mutant APP/tau transgenic mice. Neurobiol. Dis. 20, 814–822. Selkoe D. J. (2001) Alzheimer’s disease: genes, proteins, and therapy. Physiol. Rev. 81, 741–766. Suzuki N., Cheung T. T., Cai X. D., Odaka A., Otvos L. Jr, Eckman C., Golde T. E. and Younkin S. G. (1994) An increased percentage of long amyloid beta protein secreted by familial amyloid beta protein precursor (beta APP717) mutants. Science 264, 1336–1340. Tanimukai H., Grundke-Iqbal I. and Iqbal K. (2005) Up-regulation of inhibitors of protein phosphatase-2A in Alzheimer’s disease. Am. J. Pathol. 166, 1761–1771. Trojanowski J. Q. and Lee V. M. (1995) Phosphorylation of paired helical filament tau in Alzheimer’s disease neurofibrillary lesions: focusing on phosphatases. FASEB J. 9, 1570–1576. Wittmann C. W., Wszolek M. F., Shulman J. M., Salvaterra P. M., Lewis J., Hutton M. and Feany M. B. (2001) Tauopathy in drosophila: neurodegeneration without neurofibrillary tangles. Science 293, 711–714.

 2007 The Authors Journal Compilation  2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1053–1063