hyperphosphorylated, conformationally altered, and multimeric forms of tau lead to a disruption of MT stability; however, direct
evidence is lacking in vivo. In this study, an in vivo stable isotope-mass spectrometric technique was used to measure the turnover, or
dynamicity, of MTs in brains of living animals. We demonstrated an age-dependent increase in MT dynamics in two different tau
transgenic mouse models, 3xTg and rTg4510. MT hyperdynamicity was dependent on tau expression, since a reduction of transgene
expression with doxycycline reversed the MT changes. Treatment of rTg4510 mice with the epothilone, BMS-241027, also restored MT
dynamics to baseline levels. In addition, MT stabilization with BMS-241027 had beneficial effects on Morris water maze deficits, tau
partially reversed MT hyperdynamicity. Together, these data suggest that tau-mediated loss of MT stability may contribute to disease
Tauopathies are neurodegenerative diseases in which neurons
accumulate abnormal forms of tau, including hyperphosphory-
lated, conformationally altered, mislocalized, fragmented, oligo-
meric, and fibrillar forms (Lace et al., 2007; Brunden et al., 2009;
ical hallmark of extracellular plaques composed of ?-amyloid
(A?) peptides. In AD, abnormalities in tau are believed to be
downstream of A? oligomers. In some tauopathies, particularly
frontotemporal dementia with parkinsonism type-17 (FTDP-
17), tau mutations, such as P301L, can initiate disease (Ittner et
Tau binds to and modifies microtubule (MT) dynamics
(Drechsel et al., 1992; Weissmann et al., 2009). MTs, made up of
tubulin dimers, are critical to bidirectional transport of cargo to
the synapse. Disease-associated forms of tau are hypothesized to
mal tau (Alonso et al., 1996; Mocanu et al., 2008). FTDP-17 tau
mutations also alter stabilization of MTs or interfere with the
ations in MT transport result from tau abnormalities, then an
MT-stabilizing agent that restores MT function could be thera-
peutic (Brunden et al., 2009).
MT stabilizers, such as epothilones and taxanes, are used for
On the other hand, MT stabilizers are neuroprotective for A?-
mediated toxicity (Michaelis et al., 2002; Silva et al., 2011), en-
hance neurite outgrowth (Sengottuvel et al., 2011), and reduce
deficits induced by tau-P301L transfection (Shemesh and Spira,
MT numbers, and motor function in a spinal cord tauopathy
model (Zhang et al., 2005). In spinal cord injury and optic nerve
D.M.B., N.H., G.W.C., L.B.D., L.Y., S.S., C.P., J.E.M., F.Y.L, S.-H.K., G.D.V., R.E.O., and C.F.A. are employees of
Correspondence should be addressed to Dr. Donna M. Barten, Neuroscience Drug Discovery, Bristol-Myers
TheJournalofNeuroscience,May23,2012 • 32(21):7137–7145 • 7137
crush models, local administration of paclitaxel improves axonal
et al., 2011). In addition, NAP, a peptide with possible indirect
MT-stabilizing activity (Yenjerla et al., 2010), has pathological
2009). NAP is in Phase II clinical trials for tauopathy.
BMS-241027, also called epothilone D, KOS862, dEpoB, and
CRND66, is a brain-penetrant MT stabilizer (Kolman, 2004).
doses were discontinued due to poor activity and neurological
side effects (Beer et al., 2007). At low doses in a tauopathy
model, BMS-241027 reduces axonal dystrophy, increases MT
numbers, reduces pathology, and improves axonal transport
and cognitive performance (Brunden et al., 2010; Zhang et al.,
Here we provide data supporting the hypothesis that abnor-
malities in tau lead to alterations in MT turnover in vivo. MT
hyperdynamicity was reduced following alterations of tau trans-
gene expression or with BMS-241027 treatment in a tauopathy
model. MT stabilization by BMS-241027 also reduced pathology
and cognitive decline. Phase I clinical trials with low doses of
BMS-241027 in AD patients are being initiated.
Animal handling. Mice were handled strictly according either to Bristol-
Myers Squibb (BMS), Mayo Clinic, or KineMed Animal Care and Use
Committee guidelines. Mice of only a single sex were used for most
studies and is indicated in figure legends. The n value for each study is
reported either in the figure, as individual data per mouse, or in the
legend. 3xTg mice (Oddo et al., 2003) were bred at BMS and were ob-
tained congenic in the C57BL/6J background from Mark Mattson (Na-
tional Institute on Aging, Baltimore, MD). rTg4510 mice (Santacruz et
al., 2005) were bred at the Mayo Clinic or BMS. Both 3xTg and rTg4510
lines express the 4R0N P301L mutant form of human tau (tau-P301L).
rTg4510 mice have two transgenes: a tetracycline-controlled transcrip-
tional activator (tTA) driven by the Ca2?/calmodulin-dependent pro-
tein kinase II? promoter and a tau-P301L transgene driven by a
tain the transcriptional activator alone (tTA Alone), the tau transgene
alone (Tau Alone), or neither transgene (Dble Neg). Mice were housed
with a 6:00 A.M. to 6:00 P.M. light/dark cycle and allowed free access to
ing 200 ppm doxycycline (Barten et al., 2011). BMS-241027 was dosed
intraperitoneally once weekly, in 10% ethanol, 90% water, or in 8.5%
polyethylene glycol 1000 succinate, 0.5% ethanol, 90% water. A 10?
stock solution was made in the organic phase and diluted in water just
before dosing. For pathological analyses, mice were killed, and the brain
2H2O labeling and MT dynamics assay. This assay was performed as
described previously (Fanara et al., 2010) with minor modifications. In
brief, mice received an intraperitoneal bolus of 30–35 ml/kg2H2O (99.9
mol%) in saline and were then maintained on 8%2H2O in drinking
to collect plasma for body water enrichment analysis and removal of
the brain. A 2 mm slice of cortex was dissected above the ventral
hippocampus. In one study, a 2 mm slice of cerebellum was collected
as well. The tissue was placed in 2 ml of microtubule-stabilizing buf-
fer (MSB) at room temperature. MSB contained 46 mM 1,4-
piperazinediethanesulfonic acid (PIPES) buffer, pH 6.8, 0.6 mM
magnesium chloride (MgCl2), 1 mM EGTA, 10% glycerol, 0.2 M sucrose,
1 mM GTP, and protease and phosphatase inhibitors. The tissue was
The homogenate was centrifuged at 8500 ? g for 10 min at 25°C. The
a Beckman Optima TLX ultracentrifuge with a TLA100.3 rotor with the
brake turned off, or in a Beckman Optima LE-80 ultracentrifuge with Ti
50.4 rotor at 172,000 ? g for 40 min with the brake off and the deceler-
of the procedure, taking care that the MTs were not pelleted too hard or
subjected to hard braking. The pellet, containing the MT fraction, was
gently resuspended in 80 mM PIPES, pH 6.8, 1 mM MgCl2at room tem-
perature, then frozen in liquid nitrogen.
mM GTP, and phosphatase and kinase inhibitors were added. Tau-
Chemicals)-conjugated beads, followed by isolation of the MAP-2 asso-
unbound MTs have previously been shown to be largely cold stable MTs
(Fanara et al., 2010).
Tubulin was purified from these immunoprecipitation extracts using
size exclusion chromatography (Biosep SEC-S2000 column, Phenome-
nex). Tubulin samples were hydrolyzed with 6N hydrogen chloride at
110°C for 16 h. Amino acids were derivatized to form pentafluorobenzyl
derivatives for gas chromatography/mass spectrometry analysis and2H-
incorporation into alanine released from total tubulin protein was mea-
sured.2H-enrichment was calculated as the percentage increase, over
natural abundance, in the percentage of alanine derivative present as the
(M?1) mass isotopomer (EM1). The2H2O body water enrichment for
each animal was determined from plasma to allow the final calculation
of the fraction of2H-labeled tubulin in the MT preps. Measurement of
2H2O enrichment in body water was measured using a modification of
previously described procedures (Lis et al., 2008). Briefly, plasma sam-
ples were centrifuged at 500 ? g for 10 min at 4°C to eliminate cells and
other insoluble material. Aliquots of 1 ml were collected in cryogenic
Aliquots of plasma were freshly thawed on ice, diluted 1:100, and placed
into the caps of inverted sealed screw-capped vials for overnight distilla-
reaction with calcium carbide. Acetylene samples were then analyzed
using a Series 3000 cycloidal mass spectrometer (Monitor Instruments),
and M1) and calibrated against a standard curve prepared by mixing
99.9%2H2O with unlabeled water.
We observed that the tau-associated MT fraction from rTg4510 mice
fractions. Studies in which the average coefficient of variability within
groups was ?30% were considered failed. In those failed studies, the
ished or absent. In 16 studies where tau-associated MTs were measured
from rTg4510 mice, 6 failed (38%). MAP-2-associated MTs were mea-
sured from 11 studies, and only 1 (9%) failed. In three studies where
tau-associated MT variability was high, two had acceptable MAP-2-
associated MT variability. Control experiments were performed with
MTs from tau knock-out mice, and no detectable signal was observed in
the MT fraction isolated with the Tau-5 immunoprecipitation. We sus-
pect this variability in tau MTs, which has not been observed in other
transgenic mouse lines, was due to lower stability of those MTs during
the isolation procedure from the very high expression of tau in the
rTg4510 mice (Fig. 1).
Western blot analyses. Western blot analyses were performed as de-
scribed previously (Fanara et al., 2007, 2010) using Tau5 to detect all
forms of tau, ab5392 to detect MAP-2, and DM1A to detect ?-tubulin
(Sigma-Aldrich). Equal amounts of protein (BCA Protein Assay,
Thermo Scientific) were loaded in each lane, and quantitation of
?-tubulin was done using densitometry of film in the linear range. The
amount of tubulin in each sample was normalized to a standard curve
comprised of serial dilutions of purified tubulin (catalog #TLA238, Cy-
toskeleton) and a negative control (no tubulin), represented as 0%.
Morris water maze. Mice were trained in the Morris water maze
(MWM) for 6 d at 2.5 months of age, before any treatment. The mice
7138 • J.Neurosci.,May23,2012 • 32(21):7137–7145Barten,Fanaraetal.•MicrotubuleHyperdynamicityinTauTransgenicMice
ioral analysis using the rank scores for the probe trial annulus crossing
memory index to make sure there were no performance issues or signif-
icant memory bias between the experimental and control groups before
dosing. Dosing with BMS-241027 began 4 d after the completion of the
behavioral analysis. Mice were tested again in the MWM after the eighth
dose, at 4.5 months of age. The second round of water maze testing was
performed in another testing room to avoid carrying-over or savings
effects. Mice were acclimated to the experimental room for 2–3 d before
testing. The mice were placed in a water maze of 1.5 m diameter, with a
16-cm-diameter platform placed 0.5 cm under the surface of the water.
The water was made opaque with nontoxic white tempera paint, and the
water temperature was kept at 25°C. There were four training trials per
day of up to 90 s each, with a 10 s post-trial period on the platform after
each. Mice were placed under a heat lamp to dry after each trial. The
was tracked using HVS Image Advanced Tracker HVA2020 software.
Acquisition training was performed for 5 consecutive days, with a 1 min
probe trial 18 h after the last training trial. During probe trials, the plat-
form was removed from the pool. Swim speed and float time (swim
drug treatment (data not shown).
istry on paraffin-embedded sections was performed as described previ-
ously (Barten et al., 2011). AT8 and AT180 were obtained from Thermo
bert Einstein College of Medicine, Bronx, NY), anti-GFAP polyclonal
antibody from Dako, and anti-Iba1 polyclonal antibody from Wako. A
subjective scoring system was developed for each tau antibody based on
neuronal soma and dendrites. The pattern differed for each antibody.
Four sections were scored for each animal with blinding to treatment.
Silver staining was performed using the Gallyas method. A semiquanti-
of silver-positive cells per section, analyzing four sections per animal.
Silver stain scoring was based on the numbers of positive neurons, not
sections were incubated in 5% periodic acid for 5 min, washed in water,
and then placed in alkaline silver iodide solution (containing 1% silver
nitrate) for 1 min. The sections were then washed in 0.5% acetic acid for
10 min, placed in developer solution for 15 min, before washing with
0.5% acetic acid, then water. The sections were then treated with 0.1%
gold chloride for 5 min before washing in water, and incubation in 1%
sodium thiosulphate (hypo) for 5 min, before a final wash and counter-
staining with 0.1% nuclear fast red.
Hippocampal cell quantitation. Brains were cut sagittally between the
midline and ?0.75 mm laterally at 5 ?m. Nissl-stained slides were
scanned and digitized using the Aperio ScanScope. Images of the entire
brain section were captured at high resolution and were stored as files
within Spectrum (Aperio Technologies). To process images, a region of
4000 ? 4000 pixels, including the entire hippocampus, was captured
using the extract tool and saved as a JPEG file for importing into Meta-
of boxes over the hippocampus were counted. The same regions were
used for every image, and five sections were counted per animal, five
Statistical analyses. Statistical analyses were performed with ANOVA
followed by Dunnett’s multiple-comparison test using Graphpad Prism
5, unless indicated otherwise. For analysis of semiquantitative histo-
pathological scores, the Kruskal–Wallis test was used. Statistical analyses
of data with two or fewer samples per group were performed using
ANOVA with a Bonferroni’s correction and pooled estimate of error
using SAS, version 9.1. MWM was analyzed by two-way ANOVA fol-
lowed by Tukey’s post hoc tests using Sigmaplot version 11.
MT dynamics can be measured in vivo using a novel deuterium
labeling procedure (Fanara et al., 2004, 2007, 2010). Following
labeling of body fluids with2H2O, deuterium becomes incor-
porated into nonessential amino acids and then into newly
synthesized proteins. The kinetics and extent of
incorporation into soluble tubulin dimers and exchangeable tu-
bulin polymers will be equal in most cells due to the very rapid
equilibrium between these tubulin pools (Fanara et al., 2004).
However, in neurons, the dynamic exchange of newly synthe-
sized2H-tubulin dimers into MTs is reduced due to higher MT
stability (Fanara et al., 2004, 2010). Using this method, MT dy-
namics were measured in the cortex of rTg4510 mice. In these
studies, MTs were purified to three fractions: the tau-associated
MTs (normally axonal), MAP2-associated MTs (normally den-
dritic), and the unbound fraction, which mostly represents cold
stable MTs (normally in the axon hilus) (Fanara et al., 2010).
Even though there is a 13-fold overexpression of tau in rTg4510
no cross-contamination of tau or MAP-2 between fractions (Fig.
1A). Interestingly, there were no differences in the amount of
?-tubulin measured in the dimer or MT fractions between tTA
Alone and rTg4510 mice (Fig. 1B,C). Even so, there was an age-
dependent increase in the dynamic exchange of
dimers into MTs in cortex of rTg4510 mice in all MT fractions
compared with mice containing only the transactivator gene
(tTA Alone), the tau gene (Tau Alone), or neither (Dble Neg)
(Fig. 2D). Tau-P301L transgene expression can be reduced 75%
in rTg4510 mice with doxycycline treatment (Santacruz et al.,
Barten,Fanaraetal.•MicrotubuleHyperdynamicityinTauTransgenicMice J.Neurosci.,May23,2012 • 32(21):7137–7145 • 7139
immunohistochemical staining of tau
(Fig. 3A). Doxycycline treatment for 3
months in rTg4510 mice reduced MT hy-
Because rTg4510 is a very aggressive
tauopathy model, we also examined 3xTg
mice, which have a much slower and
milder phenotype. Measurements were
made from the hippocampus, where the
phospho-tau (p-tau) pathology is con-
centrated. 3xTg MTs were hyperdy-
at 20 months of age (Fig. 2E), and were
not significantly altered at any age in the
MAP-2-associated fraction (data not
shown). The unbound fraction was only
measured at 20 months of age (n ? 2),
whereitwassignificantly increased from
an average of 4.3 to 8.9% newly synthe-
sized tubulin for C57BL/6 and 3xTg
transgene expression and pathology with
altered MT turnover.
BMS-241027, a brain-penetrant MT sta-
bilizer with a long half-life in brain and a
short half-life in blood (Andrieux et al.,
2006; Brunden et al., 2010), was adminis-
tered to rTg4510 mice. BMS-241027 has
previously been used to reduce tumor formation in nude mice
with 15–35 mg/kg, i.p., treatment every other day for five cycles
Because the compound has a brain half-life of ?48 h in mice
brain with every-other-day dosing. We chose cumulative doses
that were ?10- and 100-fold lower than oncology doses by ad-
ministering the compound once weekly at 1 and 10 mg/kg, i.p.
high tau-P301L transgene expression and pathology, hyperdy-
namic tau-associated, MAP-2-associated and unbound MTs
were reduced in a dose-dependent manner to the level observed
in tTA Alone animals (Fig. 4A–C). While there was a 350% ele-
a 30% increase in cerebellum, consistent with very low levels of
transgene expression there. BMS-241027 treatment in this study
reduced MT dynamicity to baseline levels, and no lower, in both
brain regions. This suggests that BMS-241027 had a preferential
beneficial effect on hyperdynamic MTs (Fig. 4D). rTg4510 mice
were treated with BMS-241027 for shorter time periods to deter-
mine when reductions in MT hyperdynamicity in cortex would
be observed. Surprisingly, we did not observe reductions in hy-
repeatedly observed changes following 11 weeks of treatment
(Fig. 5). These studies demonstrated that BMS-241027 only al-
tered exchange of2H-tubulin dimers into MTs following an ex-
tended treatment period.
rTg4510 mice are known to develop cognitive deficits between
the ages of 2.5 and 4.5 months of age (Santacruz et al., 2005).
Because the onset of these deficits can vary, mice were tested in
the MWM at 2.5 months of age to screen for group bias before
dosing (Fig. 6A). The mice were tested again after 2 months of
the hidden platform, while rTg4510 mice treated with 1 mg/kg
BMS-241027 were able to learn the location of the hidden plat-
mg/kg BMS-241027 showed significantly more time in the target
quadrant than the vehicle-treated mice, indicating that they re-
membered the location of the platform (Fig. 6C). The vehicle-
treated animals did not show any preference for the target
quadrant, spending equal amounts of time in all quadrants.
Following behavioral testing, the mice were treated until they
were 5.5 months of age, when severe neurodegeneration is
observed in the hippocampus. An age-dependent increase of
intraneuronal p-tau accumulation, tangle formation, and
neurodegeneration has been well described in rTg4510 mice
(Ramsden et al., 2005; Santacruz et al., 2005; Spires et al., 2006).
The 1 mg/kg BMS-241027 treatment group showed a significant
reduction in tau pathology, as measured by immunohistochem-
ical staining for AT8 (p-tau 202 ? 205), AT180 (pTau 231), and
p-tau 181, with a trend for the conformation-specific antibody
stain, was also reduced in the 1 mg/kg treatment group in multi-
7140 • J.Neurosci.,May23,2012 • 32(21):7137–7145 Barten,Fanaraetal.•MicrotubuleHyperdynamicityinTauTransgenicMice
ple brain regions, with little or no benefit in the 10 mg/kg group
(Figs. 7, 8). In addition, there were significantly more CA1 and
CA3 neurons in the 1 mg/kg treatment group, similar to the
numbers measured in nontransgenic control mice (Fig. 8C).
In these pathological measures, a greater treatment effect was
observed in the 1 mg/kg group than the 10 mg/kg group. This
suggests that there may be a U-shaped dose–response curve for
some of the endpoints measured in this study. No toxicity was
observed (weight loss, or behavioral or histopathological toxic-
ity) in either 1 or 10 mg/kg groups in 2 and 3 month dosing
studies in these mice, suggesting that overt toxicity was not re-
sponsible for an apparent U-shaped dose–response curve.
It is difficult to measure MT dynamics in the brain, where MTs
are more abundant and more stable than in any other tissue. We
could not measure reproducible changes of MT mass or total or
P301L transgene expression or BMS-241027 treatment (Fig. 1;
tau hyperphosphorylation in vivo (Planel et al., 2008) and maxi-
mal tubulin acetylation in brain tissue (Brunden et al., 2011).
Interestingly, Brunden et al. (2011) were able to measure in-
creases in tubulin acetylation in the optic nerves of mice follow-
ing acute BMS-241027 treatment. Although small decreases in
the number of MTs as measured by electron microscopy and
inhibition of transport function in nerves have been described in
tauopathy models (Zhang et al., 2005, 2012), alterations in MT
turnover due to tau abnormalities have not been previously de-
MTs in vivo than control mice. Thus, tau
pathology is associated with a larger pool
of dynamic MTs in brain. This result
would not have necessarily been pre-
dicted. Although tau-P301L is less effi-
cient at binding to MTs than endogenous
in neuronal cultures (Dayanandan et al.,
1999). Therefore, although not observed,
Abnormal MT turnover in the two
transgenic lines was generally consistent
with the level of tau pathology not just
total tau protein levels. 3xTg mice have
approximately twofold overexpression of
the tau-P301L transgene and much less
pathology than the 13-fold overexpress-
ing rTg4510 line (Santacruz et al., 2005;
Barten et al., 2011). Even so, Tau Alone
transgenic mice have leaky transgene ex-
pression from the TRE promoter that is
close to the amount observed in 3xTg
mice. Tau Alone mice do not have either
2011) or hyperdynamic MTs (Fig. 2D).
Furthermore, doxycycline treatment of
rTg4510 mice completely reversed MT
hyperdynamicity (Fig. 3) and greatly re-
assay. This is the only study included where the average coefficient of variation was?30%.
Methods. B, MAP-2-associated MT results are representative of all three studies. C, D, Cold stable MTs from cortex and tau-
Barten,Fanaraetal.•MicrotubuleHyperdynamicityinTauTransgenicMiceJ.Neurosci.,May23,2012 • 32(21):7137–7145 • 7141
duced the accumulation of AT8? neurons (Fig. 2A), even
though soluble tau protein remains higher than that of very old
3xTg mice (Santacruz et al., 2005; Barten et al., 2011). Although
amyloid precursor protein and presenilin-1 mutant protein ex-
pression to the tau-P301L expression may also have contributed
accumulation and AT8? neurons between 17 and 23 months in
3xTg mice (Barten et al., 2011), which may contribute to the
significant increase in MT dynamics observed here at 20 months
(Fig. 2E). Together, these results suggest that pathological forms
of tau are associated with changes in MT turnover in vivo.
There are multiple possible mechanisms whereby pathologi-
is possible that high concentrations of human tau-P301L inter-
and STOP (stable tubule only peptide), to stabilize MTs as has
been described in vitro (Alonso et al., 1996; Mocanu et al., 2008).
of proteins now known to regulate MTs at their tip ends (Jiang
and Akhmanova, 2011). Tau phosphorylation helps determine
the spacing between microtubules, and so alterations could
influence the transport of larger cargo, such as mitochondria
been associated with disruption of neurofilament structure, loss
of dendritic spines (Zempel and Mandelkow, 2011), and altera-
tions in AMPA receptor trafficking (Hoover et al., 2010). Tau is
known to be involved in the transport of Fyn to the dendrites
where it plays a role in NMDA receptor cycling (Ittner and Gotz,
2011) and in BDNF signaling (Chen et al., 2012). This and other
roles of tau in cell signaling (Morris et al., 2011) could indirectly
lead to changes in MT dynamics, including for MAP-2 and cold
stable MTs (Job et al., 1983; Montoro et al., 1993; Fanara et al.,
2010). It is also possible that pathological forms of tau lead to a
reduction in the average length of MTs, which would then result
in more free MT ends, and more opportunities to exchange with
the labeled tubulin dimers. This could occur if MT severing en-
of abnormal tau. In fact, normal tau has been shown to protect
MTs against katanin cleavage (Qiang et al., 2006). It is likely that
the accumulation of abnormal forms of tau will have pleiotropic
effects in neurons.
dynamicity was observed in both the cortex, where transgene
expression and hyperdynamicity were high, and the cerebellum,
where transgene expression and hyperdynamicity were low.
These data suggest that hyperdynamic MTs are preferentially af-
fected by BMS-241027 at low doses. In contrast, oncologic doses
of paclitaxel are able to reduce
et al., 2004). BMS-241027 reduced MT hyperdynamicity in
rTg4510 mice with a typical dose–response curve and a maximal
effect near the 10 mg/kg dose.
On the other hand, the optimal BMS-241027 dose to reduce
icant benefit at 10 mg/kg. Only partial reduction of hyperdy-
namic MTs was observed in rTg4510 mice at the 1 mg/kg dose.
MT-stabilizing agents are known to influence MT dynamics at
molar ratios as low as 1:1000 through allosteric effects, so it is
reasonable to expect pharmacological activity at much lower
doses than that required for inhibition of mitosis (Derry et al.,
1995). It is not surprising that an apparent U-shaped dose re-
range of MT stabilization for maximal neuronal function. These
data were consistent with the “dosage effect” model previously
described for MT support of long-term viability (Feinstein and
been described for paclitaxel in the stimulation of nuclear trans-
and transport deficits induced by tau-P301L transfection
(Shemesh and Spira, 2011). Although it may be expected that
optimal neuronal function would be achieved when MT dynam-
ics in rTg4510 neurons were reduced to the same level as control
mice (as at 10 mg/kg), those neurons may differ in their optimal
set point due to the transgene expression. There was no evidence
that loss of activity with the 10 mg/kg dose was due to toxicity.
2H-label incorporation below
8 weeks of treatment. n ? 5–9 per group, and sexes are mixed. Data from 2 and 11 week
ment, the 1 mg/kg treatment group was significantly different from vehicle control in the
acquisition phase. C, Both 1 and 10 mg/kg treatment groups showed significant differences
from vehicle in the probe trial. TQ, Target; AR, adjacent right; AL, adjacent left; OP, opposite
any sex differences in the data. Data were analyzed by two-way ANOVA with significant
rants and treatment for probe data. Treatment differences are indicated on the plots.
7142 • J.Neurosci.,May23,2012 • 32(21):7137–7145Barten,Fanaraetal.•MicrotubuleHyperdynamicityinTauTransgenicMice
The pathological and functional benefits observed with 1
mg/kg BMS-241027 in rTg4510 mice are consistent with results
obtained in the PS19 tauopathy model. Epothilone D (BMS-
241027) at 1–3 mg/kg reduces dystrophic neurites, restores MT
numbers, and improves performance in the Barnes maze in a 3
month prevention trial (Brunden et al., 2010). Subsequent stud-
in a 3 month interventional trial (Zhang et al., 2012). Epothilone
D-treated mice had improvements in fast axonal transport in
neurodegeneration. We performed a study in a third tauopathy
model by treating 3xTg mice with 0.1, 1, and 10 mg/kg BMS-
241027 from 6 to 12 months of age. Although previous studies
with the BMS colony of 3xTg mice had demonstrated deficits in
MWM at 12 months, in this study the 3xTg mice performed
significantly better than vehicle-treated wild-type control mice.
In addition, there was no difference in performance between ve-
hicle and BMS-241027 treated 3xTg mice (data not shown).
the absence of a disease state.
An unexpected finding was that 3
months of BMS-241027 treatment were
required to reduce the elevation in MT
treatment was without effect on MT dy-
namics as measured using this assay, even
though cognitive benefits were observed
after 2 months and at the same age. This
finding was even more surprising because
changes in MT turnover can be measured
in response to a fear-conditioning proto-
col within 24 h with this method (Fanara
et al., 2010). In addition, increases in
inga single dose
(Brunden et al., 2011). A likely explana-
tion is that low doses of BMS-241027
cause small changes in MT stability that
do not translate to measurable changes in
the pool of dynamic MTs measured with
this assay. The changes eventually ob-
served following 3 months of dosing are
likely due to compensatory changes in MTs that can then be
measured with the stable isotope-labeling technique. There are
several scenarios consistent with this idea. In one hypothesis, tau
dysfunction leads to an accumulation of shorter MTs, and these
smaller MTs exist in nonparallel orientations that hinder trans-
port, as has been observed in Aplysia neurons transfected with
tau-P301L (Shemesh and Spira, 2010). With time, these shorter
MTs would eventually be replaced by longer and properly ori-
ented MTs. A second scenario could involve an direct anti-
inflammatory effect in glia, which would then help neurons to
cope better with the stress of overexpressing high levels of tau-
P301L. The MTs measured in this assay are predominantly neu-
ronal due to the specific isolation of MTs with tau and MAP-2
consistent with the anti-inflammatory response reported in two
2011; Sengottuvel et al., 2011). A third scenario could involve
stabilization of a small population of MTs that turn over so rap-
idly that they cannot be isolated or measured within the larger
pool of MTs measured with this assay. MTs have been imaged
moving in and out of dendritic spines in the seconds to minutes
in response to activity (Hoogenraad and Akhmanova, 2010).
Finally, BMS-241027 may be improving neuronal function
through an influence on signaling pathways, protein clear-
ance, or mitochondrial function, which would then lead to
later improvements in MT dynamics (Michaelis et al., 2005;
Shemesh et al., 2008; Vossel et al., 2010; Gardiner et al., 2011;
Lee et al., 2011; Silva et al., 2011).
port, has also been measured in a model of amyotrophic lateral
2007). No changes in total tubulin or MTs are observed, demon-
in MT turnover would not have been detected without stable
isotope labeling. Treatment of these mice for 3 months with
Gallyas silver stain. C, Quantitation of neuronal number in hippocampus. *p ? 0.05, **p ?
Barten,Fanaraetal.•MicrotubuleHyperdynamicityinTauTransgenicMiceJ.Neurosci.,May23,2012 • 32(21):7137–7145 • 7143
noscapine, a microtubule modulator, is associated with im-
rodegeneration. These data suggest that tau alterations are not
the only mechanism that can induce an increase in MT turnover
in vivo and that MT stabilization may have broader utility in
In summary, we have demonstrated that MT hyperdynamic-
ity is associated with the accumulation of pathological forms of
gene expression or stabilized MTs were able to restore normal
MT dynamics. When MTs were stabilized at very low doses, re-
ductions in tau pathology, neuronal loss, and cognitive deficits
useful as a therapeutic approach for the treatment of AD and
Alonso AC, Grundke-Iqbal I, Iqbal K (1996) Alzheimer’s disease hyper-
phosphorylated tau sequesters normal tau into tangles of filaments and
disassembles microtubules. Nat Med 2:783–787.
AndrieuxA,SalinP,SchweitzerA,Be ´gouM,PachoudB,BrunP,Gory-Faure ´
S, Kujala P, Suaud-Chagny MF, Ho ¨fle G, Job D (2006) Microtubule
stabilizer ameliorates synaptic function and behavior in a mouse model
for schizophrenia. Biol Psychiatry 60:1224–1230.
Barten DM, Cadelina GW, Hoque N, DeCarr LB, Guss VL, Yang L, Sankara-
narayanan S, Wes PD, Flynn ME, Meredith JE, Ahlijanian MK, Albright
CF (2011) Tau transgenic mice as models for cerebrospinal fluid tau
biomarkers. J Alzheimers Dis 24 [Suppl 2]:127–141.
Beer TM, Higano CS, Saleh M, Dreicer R, Hudes G, Picus J, Rarick M, Feh-
renbacher L, Hannah AL (2007) Phase II study of KOS-862 in patients
with docetaxel. Invest New Drugs 25:565–570.
Brunden KR, Trojanowski JQ, Lee VM (2009) Advances in tau-focused
drug discovery for Alzheimer’s disease and related tauopathies. Nat Rev
Drug Discov 8:783–793.
MJ, Xie SX, Ballatore C, Smith AB 3rd, Lee VM, Trojanowski JQ (2010)
tion in a transgenic mouse model of tauopathy. J Neurosci
Trojanowski JQ, Smith AB 3rd, Lee VM (2011) The characterization of
microtubule-stabilizing drugs as possible therapeutic agents for Alzhei-
mer’s disease and related tauopathies. Pharmacol Res 63:341–351.
Chen Q, Zhou Z, Zhang L, Wang Y, Zhang YW, Zhong M, Xu SC, Chen CH,
LiL,YuZP (2012) Tauproteinisinvolvedinmorphologicalplasticityin
JP, Anderton BH, Hutton M, Lovestone S (1999) Mutations in tau re-
phorylation. FEBS Lett 446:228–232.
Dehmelt L, Halpain S (2005) The MAP2/Tau family of microtubule-
associated proteins. Genome Biol 6:204.
DerryWB,WilsonL,JordanMA (1995) Substoichiometricbindingoftaxol
suppresses microtubule dynamics. Biochemistry 34:2203–2211.
DrechselDN,HymanAA,CobbMH,KirschnerMW (1992) Modulationof
protein tau. Mol Biol Cell 3:1141–1154.
Fanara P, Turner S, Busch R, Killion S, Awada M, Turner H, Mahsut A,
Laprade KL, Stark JM, Hellerstein MK (2004) In vivo measurement of
microtubule dynamics using stable isotope labeling with heavy water.
Effect of taxanes. J Biol Chem 279:49940–49947.
Fanara P, Banerjee J, Hueck RV, Harper MR, Awada M, Turner H, Husted
KH, Brandt R, Hellerstein MK (2007) Stabilization of hyperdynamic
microtubules is neuroprotective in amyotrophic lateral sclerosis. J Biol
(2010) Changes in microtubule turnover accompany synaptic plasticity
and memory formation in response to contextual fear conditioning in
mice. Neuroscience 168:167–178.
FeinsteinSC,WilsonL (2005) Inabilityoftautoproperlyregulateneuronal
mediate neuronal cell death. Biochim Biophys Acta 1739:268–279.
Gardiner J, Overall R, Marc J (2011) The microtubule cytoskeleton acts as a
key downstream effector of neurotransmitter signaling. Synapse
GreberUF,FojoT (2002) Enhancedmicrotubule-dependenttrafficking
and p53 nuclear accumulation by suppression of microtubule dynamics.
Proc Natl Acad Sci U S A 99:10855–10860.
Gozes I, Stewart A, Morimoto B, Fox A, Sutherland K, Schmeche D (2009)
Addressing Alzheimer’s disease tangles: from NAP to AL-108. Curr Alz-
heimer Res 6:455–460.
LC, Strikis D, Lemmon V, Bixby J, Hoogenraad CC, Bradke F (2011)
Microtubule stabilization reduces scarring and causes axon regeneration
after spinal cord injury. Science 331:928–931.
Hoogenraad CC, Akhmanova A (2010) Dendritic spine plasticity: new reg-
ulatory roles of dynamic microtubules. Neuroscientist 16:650–661.
Hoover BR, Reed MN, Su J, Penrod RD, Kotilinek LA, Grant MK, Pitstick R,
Carlson GA, Lanier LM, Yuan LL, Ashe KH, Liao D (2010) Tau mislo-
calization to dendritic spines mediates synaptic dysfunction indepen-
dently of neurodegeneration. Neuron 68:1067–1081.
IttnerA,KeYD,EerselJ,GladbachA,Go ¨tzJ,IttnerLM (2011) Briefupdate
on different roles of tau in neurodegeneration. IUBMB Life 63:495–502.
Ittner LM, Gotz J (2011) Amyloid-beta and tau—a toxic pas de deux in
Alzheimer’s disease. Nat Rev Neurosci 12:65–72.
Jiang K, Akhmanova A (2011) Microtubule tip-interacting proteins: a view
from both ends. Curr Opin Cell Biol 23:94–101.
Job D, Rauch CT, Fischer EH, Margolis RL (1983) Regulation of microtu-
bule cold stability by calmodulin-dependent and -independent phos-
phorylation. Proc Natl Acad Sci U S A 80:3894–3898.
Kolman A (2004) Epothilone D (Kosan/Roche). Curr Opin Investig Drugs
Lace GL, Wharton SB, Ince PG (2007) A brief history of tau: the evolving
view of the microtubule-associated protein tau in neurodegenerative dis-
eases. Clin Neuropathol 26:43–58.
Lee S, Sato Y, Nixon RA (2011) Lysosomal proteolysis inhibition selectively
disrupts axonal transport of degradative organelles and causes an
Alzheimer’s-like axonal dystrophy. J Neurosci 31:7817–7830.
Lis G, Wassenaar LI, Hendry MJ (2008) High-precision laser spectroscopy
D/H and 18O/16O measurements of microliter natural water samples.
Anal Chem 80:287–293.
Lu Q, Wood JG (1993) Functional studies of Alzheimer’s disease tau pro-
tein. J Neurosci 13:508–515.
MannaT,ThrowerDA,HonnappaS,SteinmetzMO,WilsonL (2009) Reg-
phorylated stathmin. J Biol Chem 284:15640–15649.
Michaelis ML, Chen Y, Hill S, Reiff E, Georg G, Rice A, Audus K (2002)
Amyloid peptide toxicity and microtubule-stabilizing drugs. J Mol Neu-
Michaelis ML, Seyb KI, Ansar S (2005) Cytoskeletal integrity as a drug tar-
get. Curr Alzheimer Res 2:227–229.
Mocanu MM, Nissen A, Eckermann K, Khlistunova I, Biernat J, Drexler D,
Petrova O, Scho ¨nig K, Bujard H, Mandelkow E, Zhou L, Rune G, Man-
delkowEM (2008) Thepotentialfor?-structureintherepeatdomainof
tau protein determines aggregation, synaptic decay, neuronal loss, and
coassembly with endogenous tau in inducible mouse models of tauopa-
thy. J Neurosci 28:737–748.
Montoro RJ, Díaz-Nido J, Avila J, Lo ´pez-Barneo J (1993) N-methyl-D-
protein 2 and potentiates excitatory synaptic pathways in the rat hip-
pocampus. Neuroscience 54:859–871.
Morris M, Maeda S, Vossel K, Mucke L (2011) The many faces of tau. Neu-
ate R, Mattson MP, Akbari Y, LaFerla FM (2003) Triple-transgenic
and synaptic dysfunction. Neuron 39:409–421.
Planel E, Krishnamurthy P, Miyasaka T, Liu L, Herman M, Kumar A, Brette-
ville A, Figueroa HY, Yu WH, Whittington RA, Davies P, Takashima A,
Nixon RA, Duff KE (2008) Anesthesia-induced hyperphosphorylation
7144 • J.Neurosci.,May23,2012 • 32(21):7137–7145Barten,Fanaraetal.•MicrotubuleHyperdynamicityinTauTransgenicMice
detaches 3-repeat tau from microtubules without affecting their stability
in vivo. J Neurosci 28:12798–12807.
Qiang L, Yu W, Andreadis A, Luo M, Baas PW (2006) Tau protects micro-
tubules in the axon from severing by katanin. J Neurosci 26:3120–3129.
Ramsden M, Kotilinek L, Forster C, Paulson J, McGowan E, SantaCruz K,
Guimaraes A, Yue M, Lewis J, Carlson G, Hutton M, Ashe KH (2005)
Age-dependent neurofibrillary tangle formation, neuron loss, and mem-
ory impairment in a mouse model of human tauopathy (P301L). J Neu-
A, DeTure M, Ramsden M, McGowan E, Forster C, Yue M, Orne J, Janus
C,MariashA,KuskowskiM,HymanB,HuttonM,AsheKH (2005) Tau
suppression in a neurodegenerative mouse model improves memory
function. Science 309:476–481.
Sengottuvel V, Leibinger M, Pfreimer M, Andreadaki A, Fischer D (2011)
Taxol facilitates axon regeneration in the mature CNS. J Neurosci
Hasegawa M, Hisanaga S (2012) Regulation of mitochondrial transport
and inter-microtubule spacing by tau phosphorylation at the sites hyper-
phosphorylated in Alzheimer’s disease. J Neurosci 32:2430–2441.
Shemesh OA, Spira ME (2010) Hallmark cellular pathology of Alzheimer’s
disease induced by mutant human tau expression in cultured Aplysia
neurons. Acta neuropathologica 120:209–222.
Shemesh OA, Spira ME (2011) Rescue of neurons from undergoing hall-
mark tau-induced Alzheimer’s disease cell pathologies by the antimitotic
drug paclitaxel. Neurobiol Dis 43:163–175.
ShemeshOA,ErezH,GinzburgI,SpiraME (2008) Tau-inducedtrafficjams
ing. Traffic 9:458–471.
Silva DF, Esteves AR, Arduino DM, Oliveira CR, Cardoso SM (2011)
Amyloid-beta-induced mitochondrial dysfunction impairs the au-
tophagic lysosomal pathway in a tubulin dependent pathway. J Alzhei-
mers Dis 26:565–581.
Spires TL, Orne JD, SantaCruz K, Pitstick R, Carlson GA, Ashe KH, Hy-
man BT (2006) Region-specific dissociation of neuronal loss and
neurofibrillary pathology in a mouse model of tauopathy. Am J Pathol
Vossel KA, Zhang K, Brodbeck J, Daub AC, Sharma P, Finkbeiner S, Cui B,
Mucke L (2010) Tau reduction prevents Abeta-induced defects in ax-
onal transport. Science 330:198.
Weissmann C, Reyher HJ, Gauthier A, Steinhoff HJ, Junge W, Brandt R
(2009) Microtubule binding and trapping at the tip of neurites regulate
tau motion in living neurons. Traffic 10:1655–1668.
L (2010) The neuroprotective peptide NAP does not directly affect po-
lymerization or dynamics of reconstituted neural microtubules. J Alzhei-
mers Dis 19:1377–1386.
Zempel H, Mandelkow EM (2011) Linking amyloid-? and tau: amyloid-?
eton. Neurodegener Dis. Advance online publication. Retrieved April 11,
Zhang B, Maiti A, Shively S, Lakhani F, McDonald-Jones G, Bruce J, Lee EB,
Xie SX, Joyce S, Li C, Toleikis PM, Lee VM, Trojanowski JQ (2005)
Microtubule-binding drugs offset tau sequestration by stabilizing micro-
tubules and reversing fast axonal transport deficits in a tauopathy model.
Proc Natl Acad Sci U S A 102:227–231.
SX, Ballatore C, Smith AB 3rd, Lee VM, Brunden KR (2012) The
neurotoxicity, cognitive deficits, and Alzheimer-like pathology in an in-
terventional study with aged tau transgenic mice. J Neurosci 32:3601–
Barten,Fanaraetal.•MicrotubuleHyperdynamicityinTauTransgenicMiceJ.Neurosci.,May23,2012 • 32(21):7137–7145 • 7145