• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of molcellbPermissionsJournals.ASM.orgJournalMCB ArticleJournal InfoAuthorsReviewers
Mol Cell Biol. Oct 2010; 30(19): 4626–4643.
Published online Aug 2, 2010. doi:  10.1128/MCB.01493-09
PMCID: PMC2950521

Loss of Hsp110 Leads to Age-Dependent Tau Hyperphosphorylation and Early Accumulation of Insoluble Amyloid β[down-pointing small open triangle]


Accumulation of tau into neurofibrillary tangles is a pathological consequence of Alzheimer's disease and other tauopathies. Failures of the quality control mechanisms by the heat shock proteins (Hsps) positively correlate with the appearance of such neurodegenerative diseases. However, in vivo genetic evidence for the roles of Hsps in neurodegeneration remains elusive. Hsp110 is a nucleotide exchange factor for Hsp70, and direct substrate binding to Hsp110 may facilitate substrate folding. Hsp70 complexes have been implicated in tau phosphorylation state and amyloid precursor protein (APP) processing. To provide evidence for a role for Hsp110 in central nervous system homeostasis, we have generated hsp110/ mice. Our results show that hsp110/ mice exhibit accumulation of hyperphosphorylated-tau (p-tau) and neurodegeneration. We also demonstrate that Hsp110 is in complexes with tau, other molecular chaperones, and protein phosphatase 2A (PP2A). Surprisingly, high levels of PP2A remain bound to tau but with significantly reduced activity in brain extracts from aged hsp110/ mice compared to brain extracts from wild-type mice. Mice deficient in the Hsp110 partner (Hsp70) also exhibit a phenotype comparable to that of hsp110/ mice, confirming a critical role for Hsp110-Hsp70 in maintaining tau in its unphosphorylated form during aging. In addition, crossing hsp110/ mice with mice overexpressing mutant APP (APPβsw) leads to selective appearance of insoluble amyloid β42 (Aβ42), suggesting an essential role for Hsp110 in APP processing and Aβ generation. Thus, our findings provide in vivo evidence that Hsp110 plays a critical function in tau phosphorylation state through maintenance of efficient PP2A activity, confirming its role in pathogenesis of Alzheimer's disease and other tauopathies.

Diseases like Alzheimer's disease (AD) and other tauopathies are defined by the expression of neurofibrillary tangles (NFTs) deposited mainly in neurons. The NFTs are aggregates of the hyperphosphorylated tau (p-tau) (3, 74). Normal tau increases microtubule stability, but tau can be hyperphosphorylated under disease conditions and released from microtubules (3, 5, 6). The molecular mechanisms involved in the formation of NFTs are not completely understood. However, accumulation of abnormal p-tau and NFTs causes neurodegeneration (3). A number of protein kinases, including glycogen synthase kinase 3 (GSK3) and cyclin-dependent protein kinase 5 (CDK5), have been shown to phosphorylate tau at Thr231 and Ser262 as well as several other sites that flank the microtubule binding repeat, leading to tangles of paired helical filaments (PHFs) similar to those observed in the brains of patients with AD (54, 72). Evidence shows that GSK3 physically interacts with tau and is thought to be the main contributor to the formation of NFTs and amyloid β (Aβ) plaques in AD patients (18, 53, 54). Phosphorylation of GSK3a/b at S9/S21 which is inhibitory to its activity during insulin signaling, leads to phosphorylation of tau in neurons (80). GSK3a/b phospho-S9/S21, p-tau, and 14-3-3zeta have been isolated in a 500-kDa complex, and the interaction has been shown to result in tau phosphorylation by GSK3 (1, 80). Although not well characterized, p-tau has been shown to be dephosphorylated by the B family regulatory subunit of the heterotrimeric PP2A holoenzyme (76). There are two protein phosphatase 2A (PP2A) binding sites on microtubule tau binding repeats, perhaps allowing tau to be more efficiently dephosphorylated by PP2A catalytic subunit (76).

Both GSK3 and CDK5 are also known to be involved in the phosphorylation of amyloid precursor protein (APP) at Thr668 and APP processing and Aβ production (53, 58). Studies suggest that amyloid peptide can activate GSK3 signaling, and the increase in GSK3 activity can then contribute to abnormal APP processing. Indeed, reduction in GSK3 activity reduces amyloid peptide production in murine AD models (18, 53, 57, 71). Reduction in PP2A activity leads to altered APP regulation as well (26, 43). Additional molecules that affect tau hyperphosphorylation and APP processing are the peptidyl prolyl isomerases (9, 36, 51). Deletion of Pin1 isomerase in vivo leads to p-tau and neurodegeneration (42). Crossing Pin1-deficient mice with transgenic mice expressing mutant APP (APPβsw) leads to abnormal APP processing and accumulation of toxic amyloid β42 (Aβ42) species. Pin1, therefore, is implicated in isomerization of tau, perhaps facilitating its dephosphorylation (42). The presence of Pin1 has been implicated in promoting nonamyloidogenic processing of APP and reduction in toxic Aβ42 production (51).

Hsp70/Hsc70 has been shown to preferentially bind to a hyperphosphorylated form of tau in the diseased human brain (49). Cross talk between the ubiquitin proteasome system (UPS) and molecular chaperones might also be critical in regulating the deposition and toxicity of tau (8, 16). These results suggest that the activity of Hsp70 and Hsp90 preserve the native structure and function of tau protein. Hsp70 and the C-terminal Hsp70-interacting protein (Chip) have been shown to regulate tau ubiquitination and degradation (11, 12, 21, 52, 65). Interestingly, Chip and βAPP interact, and Chip and Hsp70/90 expression have been shown to lower the cellular levels of Aβ and reduce Aβ toxicity in vitro (39). Misfolded proteins are either degraded through the UPS or are folded, at least in part, by the Hsps (4, 7).

Eukaryotic cells possess a class of heat shock proteins (Hsps) related to the Hsp70 family. This Hsp100 family of proteins contains Hspa41 (Apg1 or OSP94), Hsp94 (Apg2), and Hsp110 (2, 17, 28, 61, 70, 77, 78). They were initially considered to be “holdases” that keep denatured proteins in solution, and no client proteins have been described for them (14, 15, 56, 62). Hsp110 interacts with Hsp70 and increases its ATPase activity (15, 56, 62). The main function of Hsp110 appears to be a nucleotide exchange factor (NEF) for Hsp70 (14, 64). In general, Hsp110 is known to induce suppression of aggregation and protein refolding, and it protects proteins from the damaging effects of various stresses; however, its physiological function in mammalian cells remains unknown (15, 60). In these studies, we examined the role of Hsp110 in central nervous system (CNS) homeostasis in vivo. We have found that hsp110/ mice exhibit an age-dependent accumulation of p-tau that is associated with pathological features, such as the appearance of NFTs and neurodegeneration. We also show that lack of Hsp110 leads to accelerated pathology as evidenced by the early appearance of senile plaques containing Aβ42 (a major toxic species [46]) in an AD transgenic mouse model. At the biochemical level, we show that Hsp110 interacts with tau, a number of Hsps, GSK3, Pin1, and PP2A. Furthermore, tau immunocomplexes pulled down from hsp110/ brain extracts possess elevated levels of PP2A, but the pulled-down PP2A has significantly lower activity than the PP2A from wild-type mice. Our studies therefore suggest a critical role for Hsp110 in maintaining the proper folding environment that is required for phosphorylation and dephosphorylation of tau and APP processing in vivo.


Generation of hsp110-deficient mice.

To generate the hsp110 targeting vector, a 129/SvJ mouse genomic DNA phage library (Lambda fixII vector; Stratagene, La Jolla, CA) was used to identify clones containing the hsp110 gene using a mouse hsp110 cDNA as a probe. This 514-bp probe spanned exons 1 to 5 and was amplified by PCR using forward (5′-GGG GGA TCC ATG TCG GTG GTT GGG CTA GAC G3′) and reverse (5′-GCA AGC AGT TCA AGC CCA CAA TCT-3′) primers. Targeting vector construction was based on a lacZ-neo-tk (pN-Z-tk2) template plasmid vector containing a β-galactosidase (lacZ) gene fragment with the bovine growth hormone poly(A) signal [lacZ-poly(A)], a neomycin resistance gene driven by the thymidine kinase (tk) promoter with the simian virus 40 poly(A) signal [tk/neo-poly(A)], and flanking tk gene cassettes (32). The tk/neo-poly(A) fragment was flanked by Cre recombinase recognition (loxP) sequences to allow removal of the selectable marker gene (Neo) from the targeted locus by intercrossing the mutant mice with transgenic mice expressing the cre recombinase gene [B6.C-Tg(CMV-cre) cgn/J (Jackson Laboratory, Bar Harbor, ME)]. In the targeting construct, Hsp110 exon 1 at ATG and 910 bp of intron 1 were deleted and replaced in frame with a 3.5-kb lacZ-neomycin cassette using a 2.5-kb insert of hsp110 at the proximal end and a 5-kb insert of hsp110 gene at the distal end. The 2.5-kb proximal region was amplified using forward (5′-ACG CGT CGA CGA TCC TTC TTA AAA ATC TAC-3′) and reverse (5′-GTA AAG CTT GGC TGG CCC GGT CCG CCT C-3′) primers and contained SalI and HindIII restriction enzyme sites. The 5-kb distal end contained restriction enzyme NotI and XhoI sites. The NotI site was generated by PCR, and the XhoI site is located downstream of exon 7 in the hsp110 gene. The DNA fragments were ligated into pN-Z-tk2. The final hsp110 targeting vector was sequenced and mapped, the vector was linearized by SalI restriction enzyme and electroporated into C57BL/6 embryonic stem (ES) cells, and G418-resistant, ganciclovir-sensitive, clones were selected. From the 135 isolated ES clones that were analyzed by Southern blotting, 20 clones contained the correctly targeted allele using a probe external to the targeting vector. Two positive ES clones were injected into C57BL/6J blastocysts, and the resulting chimeric male mice were crossed with C57BL/6J females to generate germ line transmission and an hsp110 heterozygous mouse line. The hsp110/ mice were born with the expected Mendelian frequency and were phenotypically normal and fertile for a period of at least 1 year. Experiments described in this study were approved by the Institutional Animal Care and Use Committee. The hsp110 chimeras were generated by the Medical College of Georgia Mouse Embryonic Stem Cell and Transgenesis Core Laboratory.

Mouse lines.

The hsp110/ mice were in C57BL/6 genetic background. Tg2576+ transgenic mice were purchased from Taconic (Hudson, NY) and were in C57BL/6J SJL genetic background. The Tg2576+ transgenic mice overexpress a 695-amino-acid splice form (Swedish mutation K670N M671I) of the human APPβ (29). Generation of hsp70i/ mice (C57BL/6 genetic background) will be reported elsewhere (D. Moskophidis, unpublished data).

Southern blot analysis and genotyping.

Genomic DNA was digested with BamHI and fractionated on a 0.8% agarose gel and transferred to a nylon membrane (Hybound-N+; Amersham Biosciences). The probe used for Southern blot analysis was a 730-bp fragment of Hsp110 cDNA that was PCR amplified from genomic DNA using the following primers: 5′-GGG GGA ATT CCT AGG ATG GGC AAA G-3′ and 5′-GGC GGA ATT CGA TGC CCT TTC AAG AA-3′. The probe was labeled by random priming with [α-32P]CTP using NEBlot kit (New England Biolabs, Inc., MA). Wild-type and mutant Hsp110 genomic DNA digested with BamHI generated 15.6-kb and 13-kb fragments, respectively. For routine genotyping of mice, DNA extracted from tail DNA was used in multiplex PCR analysis to verify a wild-type 242-bp fragment, and a mutant 405-bp fragments using the following primers: primer 1, 5′-ACATAAGGCTGAGCGATTGG-3′; primer 2, 5′-ATGTAGCAGCTCTGTGAGCCTAC-3′; and primer 3, 5′-CAGGAAGATCGCACTCCAG-3′. Primers 1 and 2 were located in exon 1, and primer 3 was located in the LacZ cDNA (48, 81).

Primary neuronal cell culture.

Neuronal cultures were prepared from cortices extracted from embryonic day 18 (E18) embryos as previously reported (27).

Sarkosyl-soluble and -insoluble fractions of brain tissue.

To prepare Sarkosyl-soluble and -insoluble fractions, the method of Dickey et al. (12) was used. Brain tissue was homogenized in 50 mM Tris (pH 8.0), 5 mM KCl, 274 mM NaCl buffer, and a cocktail of proteases and phosphatase inhibitor (10 mM sodium fluoride and 2 mM sodium vanadate). The homogenate was subjected to centrifugation at 60,000 rpm for 15 min, and the supernatant was the soluble S1 fraction. The pellet was further homogenized in 10 mM Tris HCl (pH 7.4), 0.8 M NaCl, 10% sucrose, and 1 mM EGTA buffer and reultracentrifuged. The supernatant from the second centrifugation step was mixed with 1% Sarkosyl and incubated for 1 h at 37°C. The second extract was subjected to ultracentrifugation for 30 min, and the supernatant was the Sarkosyl-soluble (S2 fraction). The pellet was the Sarkosyl-insoluble (P3 fraction) and was resuspended in Tris-EDTA.

Measurements of α- and β-secretase activities and Aβ production.

α-Secretase activity (catalog no. FP001) and β-secretase activity (catalog no. FP002) were measured using kits purchased from R&D (Minneapolis, MN). Mouse Aβ40 (catalog no. KMB3481), mouse Aβ42 (catalog no. KMB3441), human Aβ40 (catalog no. KHB3481), and human Aβ42 (catalog no. KHB3441) were detected using immunoassay (calorimetric) kits purchased from Invitrogen (Camarillo, CA). Preparation of soluble and insoluble fractions of brain extracts, and all procedures were conducted as suggested by the manufacturers.

Peptidyl-propyl cis-trans isomerase and PP2A activities.

Peptidyl prolyl isomerase (PPIase) activity was determined as previously described with minor modifications (67). Briefly, 200 μl of assay mixture contained 4 μl of fresh brain cell lysate (2 mg/ml), 1 μl of α-chymotrypsin (60 mg/ml; Sigma), and 193 μl of HEPES buffer (32 mM, pH 7.8). The reaction was started by adding 2 μl of substrate (25 mg/ml of N-succinyl-Ala-Ala-Pro-Phe-p-nitroanilide) (Sigma). For control groups, brain cell lysates were replaced with HEPES buffer. The reaction was monitored by measuring the absorbance at 395 nm for 0 to 10 min.

To determine PP2A activity, total tau was pulled down from 1 mg of brain extracts. Fifty microliters of buffer containing 40 mM Tris-HCl (pH 8.4), 34 mM MgCl2, 4 mM EDTA, and 4 mM dithiothreitol (DTT) was added to protein A-PP2A immunocomplexes, and the mixture was incubated with 20 mM p-nitrophenol phosphates (pNpp) substrate and incubated at 25°C for 3 h. The absorbance was measured at 405 nm (76).

Purified proteins.

The purified Hsp110 (human) was a generous gift from J. Subjeck (Roswell Park, NY). The purified tau441 was purchased from rPeptide (Bogart, GA).

Immunoprecipitation and immunoblotting.

Brain tissue was homogenized in radioimmunoprecipitation assay (RIPA) buffer containing 150 mM NaCl, 1% NP-40, 0.5% deoxycholate, 0.1% SDS, 50 mM Tris (pH 7.5), 1× cocktail of protease inhibitor, 10 mM phenylmethylsulfonyl fluoride (PMSF), 1% sodium pyrophosphate, 10 mM sodium fluoride, and 2 mM sodium vanadate. Samples were centrifuged at 10,000 × g for 20 min at 4°C. Immunoprecipitation and immunoblotting assays were performed as previously described (30). For negative controls in immunoprecipitation analyses, we used antibody against green fluorescent protein (GFP) when mouse primary antibody was used, and rabbit or goat serum when antibodies from these species were used. Primary antibodies were as follows: Hsp90, Hsc70, Hsp60, and Hsp40 (Stressgen); Pin1 and PP2A-C a/b (sc166034); pPP2A-C a/b (Tyr307) (sc12615); PP2A-B55-a (sc81606) (Santa Cruz); Hsp110 (sc-1804 and sc-6241; Santa Cruz); Chip (Cell Signaling); NeuN (Chemicon); glial fibrillary acidic protein (GFAP) (Dako); βAPP (Sigma, Saint Louis, MO) [catalog no. NAB228; this antibody recognizes human β-amyloid peptide, full-length APP, soluble APPβ′ and APPα, C99 cleavage form, and Aβ(1-40/42)]; soluble APPβsw (sAPPβsw) (catalog no. 1032; this antibody recognizes human sAPPβsw) (IBL, Japan). Tau and phospho-tau antibodies (PHF1, CP13, total tau, and MC1) were generous gift from P. Davies (Albert Einstein College of Medicine, NY). The p-tau antibody 12E8 was a generous gift from Peter Seubert (Elan Pharmaceuticals, Inc., San Francisco, CA).

Histology and immunohistochemistry.

The brains were fixed in 4% paraformaldehyde and embedded in paraffin. For immunostaining, 7-μm tissue sections were deparaffinized in xylene and rehydrated in a series of alcohol-water mixtures. Antigen retrieval was performed by placing the slides in 10 mM sodium citrate (pH 6.0) and steam for 30 min. After the slides were rinsed in phosphate-buffered saline (PBS), tissue sections were blocked in PBS containing 3% bovine serum albumin (BSA) for 2 h at 4°C. The tissue sections were incubated in primary antibody diluted in PBS containing Tween 20 (PBST) and 3% BSA for 16 h at 4°C. Antibody or antigen was detected with fluorescent Cy3-conjugated anti-mouse (or anti-rabbit) IgG secondary antibody. Nuclei were stained with 4′,6′-diamidino-2-phenylindole (DAPI) (27). For β-galactosidase staining of tissue sections, cryosections were stained as described previously (31). For Bielschowsky staining, tissue sections were deparaffinized, rehydrated, and stained with 10% silver nitrate solution for 1 h. After the tissue sections were rinsed with water, the sections were immersed in ammonium silver solution. The sections were then immersed in 1% ammonium hydroxide and then placed in ammonium silver solution that included developer (100 ml water, 20 ml 10% formalin, 50 μl concentrated nitric acid, and 0.5 g citric acid). After color formation was monitored by using a microscope, the sections were immersed in 1% ammonium hydroxide, rinsed with water, and placed in a 5% sodium thiosulfate solution for 5 min. The sections were then rinsed with water, dehydrated, cleared, and mounted. For thioflavin T staining, deparaffinized brain tissue sections were rehydrated in water. The antigens were retrieved, and the tissue sections were then washed with water and stained with 0.05% thioflavin T (catalog no. T3516; Sigma) in 50% ethanol for 8 min. The tissue sections were then differentiated in 70% ethanol for 5 min and rinsed twice with water and mounted. Immunostained sections were analyzed by Zeiss Axio Imager or a confocal microscope. For Axio Imager, the excitation and emission wavelengths used were 365 and 445/50 nm (DAPI), 470/40 and 525/50 nm (fluorescein isothiocyanate [FITC]), 546/11 and 605/75 nm (Cy3), respectively. For confocal microscopy (in Fig. Fig.1E1E only), the excitation and emission wavelengths used were 488 and 546 (FITC), and 546 and 580/90 (Cy3), respectively. For immunostaining experiments, all tissue sections were also stained with the secondary antibody only as a control for background staining.

FIG. 1.
Targeted disruption of the hsp110 gene in mice. (A) Schematic diagrams of the wild-type (WT) hsp110 locus, targeting vector, and the predicted targeted allele following homologous recombination. Exons are indicated by black boxes (E1 to E7). The location ...

Behavioral studies.

Behavioral studies were performed with wild-type and hsp110/ male mice at 12 months of age. Studies were performed and analyzed in the Small Animal Behavioral Core at the Medical College of Georgia directed by Alvin Terry (68). The tests performed were fear conditioning (context-dependent freezing and cued-dependent freezing), Y-maze/spontaneous alternation, and open-field locomotor activity (40, 45). To test for motor function, the ability of mice to balance on a 1.2-cm diameter bar fixed at a height of 40 cm was determined. The ability of hsp110-deficient mice and age-matched wild-type controls at 18 months of age to remain on the bar for 45 s was assessed. The results are expressed as the means plus standard errors of the means (SEMs).

Statistical analyses.

All experiments were performed at least three times. Three to eight mice were used for each time point and/or genotype. Multiple sections of tissues (n = 5 to 10) from each mouse were analyzed for each histological or immunohistological analysis. For statistical analyses, data were expressed as means plus standard errors of the means and unpaired two-tailed Student's t test. In some analyses, one-way analyses of variance (ANOVA) were used to compare groups as indicated in the figure legend. Differences between groups were considered significant at P < 0.05.


Hsp110 is expressed in the CNS.

To investigate the physiological role of Hsp110 in the CNS, we generated mice with targeted disruption of the hsp110 gene and replaced the gene with β-galactosidase (Fig. 1A to E). Previous reports indicate that Hsp110 is expressed in neurons in the cerebral cortex, hippocampus (Cornu ammonis 1 [CA1], CA2, CA3, and dentate gyrus [DG]), thalamus, hypothalamus, and in the Purkinje cells of the cerebellum but is absent in granule cells of the cerebellum, astrocytes, and glial cells (15, 33, 63). To confirm and extend these observations, we performed lacZ staining of different regions of the CNS in hsp110/-lacZ reporter mice. The results indicate that hsp110-lacZ is expressed in both hippocampus and cortical regions of the brain (Fig. (Fig.1B).1B). We also stained cultured neurons for the presence of lacZ (Fig. (Fig.1C).1C). The data indicate that hsp110-lacZ is expressed constitutively in the cortical neurons during embryogenesis. Using antibody to Hsp110, immunohistochemical staining shows Hsp110 expression in the pyramidal cells of the hippocampus, Purkinje cells in the cerebellum (Fig. (Fig.1D,1D, panel b, black arrow), and the majority of cortical neurons (Fig. (Fig.1D,1D, panel c) but is absent in granule cells of the cerebellum (Fig. (Fig.1D,1D, panel b, black arrowhead). Figure Figure1D,1D, panel a, shows negative control.

We performed double immunofluorescence analyses to confirm that a neuron-specific marker (NeuN) coexpresses with Hsp110 in the cortex (hippocampal region). Data presented in Fig. Fig.1E1E show that Hsp110 coexpresses in the majority of NeuN-positive cells in the cortex as well as in CA1, CA2, CA3, and DG of the hippocampus (data not presented).

hsp110−/− mice exhibit age-dependent accumulation of hyperphosphorylated tau and NFTs.

Several molecular chaperones, such as Hsp70, Hsp90, and Hsp27, have been shown to interact with various microtubule components, including tau (41). Since Hsp110 is the NEF for Hsp70, we examined the expression of tau, a microtubule-associated protein, and p-tau in the presence or absence of the Hsp110 gene. The results show that total tau expresses in neurons in both wild-type and hsp110/ brain tissue (Fig. (Fig.2A2A ). Surprisingly, a significant number of neurons also stained positively for p-tau (pS202/T205; S396/S404; T231; S262/356) in the hippocampal region of the brain from hsp110/ mice (Fig. (Fig.2A)2A) (see Fig. S1 in the supplemental material). Quantitation of the immunoreactive cells indicated a significant increase of p-tau in brain tissue from hsp110/ mice that increased with age (Fig. (Fig.2A,2A, graphs). The positive immunoreactivity for p-tau was apparent only in brain tissue from hsp110/ mice, and not in wild-type mice in all age groups. In addition to the cortex, immunoreactivity to p-tau at pS396/S404 and pS202/T205 was also observed in the corpus callosum and thalamus of hsp110/ mice (data not shown). In 4- to 6-week-old mice, p-tau (highest immunoreactivity using PHF1 and CP13 antibodies) initially appears in the corpus callosum and then extends to the thalamus and cortical areas.

FIG. 2.
Tau is hyperphosphorylated and forms NFTs in the brains of hsp110/ mice. (A) For the micrographs shown to the left of the panel, fixed and processed sections of the cortex (hippocampal region) from the brains of 24-week-old wild-type ...

Previous studies indicate that the ubiquitin ligase Chip (C-terminal Hsp70-interacting protein) participates in the clearance of p-tau (12, 35). Indeed, chip/ mice exhibit immunoreactivity to CP13 (S202/T205) and 12E8 (S262/S356) p-tau antibodies. However, chip/ mice do not exhibit any immunoreactivity to MC1 antibody, which recognizes the Alzheimer's-type conformational tau epitope (12, 35). In contrast, previous data indicate that pin1/ mice exhibit positive immunoreactivity to many phospho-specific tau antibodies, including MC1 (42). Therefore, to examine whether hsp110/ neurons show MC1 immunoreactivity, we immunostained tissue sections from the olfactory region (forebrain) of hsp110/ mice (Fig. (Fig.2A).2A). The data indicate that hsp110/ neurons exhibit positive immunoreactivity to the conformation-specific tau antibody in this brain region. No MC1 immunoreactivity was observed in the cortex of hsp110/ or wild-type mice (see Fig. S1 in the supplemental material). Quantitation of the data indicates a significant age-dependent increase in the amount of immunoreactive p-tau in brain tissue (forebrain) of hsp110/ mice compared to wild-type mice (P < 0.04) (Fig. (Fig.2A,2A, graphs).

These results indicate that the brain tissue of hsp110/ mice exhibits positive immunoreactivity to p-tau-specific antibodies and that this increases with age.

To confirm the observation made with immunohistochemical staining of p-tau, we performed biochemical analyses and detected p-tau in Sarkosyl-soluble and -insoluble fractions (12) prepared from the aging brains of hsp110/ and wild-type mice. Results indicate that the soluble (S1) and Sarkosyl-soluble (S2) fractions of hsp110/ brain extracts exhibit the presence of p-tau using phosphospecific antibodies PHF1 and CP13 in 24- to 30-week-old mice but is not present in 4- to 6-week-old mice (Fig. (Fig.2B).2B). When the mice were 24 to 30 weeks old, p-tau could also be detected in the Sarkosyl-insoluble pellet fraction (P3) using PHF1 or CP13 antibody in brain extracts from hsp110/ mice (Fig. (Fig.2B).2B). The presence of p-tau in the soluble brain extracts of hsp110/ mice was confirmed by immunoblotting using PHF1 antibody after treating cell extracts with calf intestinal alkaline phosphatase (CIP) (Fig. (Fig.2C).2C). Treatment of hsp110/ brain tissue extracts with CIP significantly reduced PHF1 immunoreactivity.

One of the hallmarks of neurodegeneration that is also evident in brains expressing p-tau is the presence of NFTs (3, 22). NFT-like structures were detectable in both cerebral cortex (Fig. (Fig.2D,2D, panels b and d) and hippocampus regions (CA1) (Fig. (Fig.2D,2D, panel c) in the brains of 24- to 32-week-old hsp110/ mice, but not in wild-type mice (cortex) (Fig. (Fig.2D,2D, panel a) using Bielschowsky staining. Thioflavin T staining of the cortex (hippocampal region) also shows the presence of NFTs (Fig. (Fig.2E,2E, bottom panel). The number of cells showing NFTs using Bielschowsky staining is presented in Fig. Fig.2F2F.

Accumulation of p-tau has been shown to induce apoptosis of neurons leading to neurodegeneration (3, 38). To determine whether the number of apoptotic cells increases with age in hsp110/ brains, we quantitated apoptotic cells present in the brain tissue sections (hippocampal region) following terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling (TUNEL) staining. Data showed an increase in the number of apoptotic cells compared to wild-type mice in an age-dependent manner (3 ± 1 in 4- to 6-week-old mice and 12 ± 0.5 in 24- to 32-week-old mice). Overall, no evidence of significant apoptosis was found in hsp110/ mice compared to wild-type mice.

hsp110−/− mice exhibit behavioral deficit.

p-tau has been shown to cause apoptosis of neurons and neurodegeneration (3, 42). Since we observed age-dependent p-tau accumulation, neuronal death, and NFTs in hsp110/ mice, we subjected 12-month-old wild-type and hsp110/ mice to contextual and cued fear conditioning and Y-maze/spontaneous alternation and open-field behavioral tests. The data presented in Fig. 3A to C indicate that hsp110/ mice exhibit a significant reduction in the contextual fear conditioning test (Fig. (Fig.3A).3A). The context-dependent freezing memory test evaluates the learned aversion of mice for an environment that has been associated with a negative stimulus (a mild shock in these experiments). The dependent variable is freezing behavior. The contextual fear learning is dependent on the hippocampus, a region of the brain that has been associated with cognitive decline in Alzheimer's disease. The cued fear conditioning test is independent of the hippocampus and was not significantly altered in hsp110/ mice compared to wild-type mice (Fig. (Fig.3A)3A) (61). The 1-year-old hsp110/ mice behaved comparably to wild-type mice in overall response to Y-maze/spontaneous alternation and open-field tests (Fig. 3B and C). The spontaneous alternation test assesses the normal navigation behavior of mice. For each mouse, the number of arms entered, distance traveled, and mean velocity were used to determine locomotor activity. Successes in this test are indicated by a high rate of alternation in the control groups, indicating that mice can remember which arm was entered last. This test evaluates short-term memory. In the case of open-field tests, animal movements were monitored by a series of horizontally and vertically mounted photo beams.

FIG. 3.
hsp110/ mice exhibit deficits in contextual fear conditioning test. (A) One-year-old wild-type and hsp110/ male mice (n = 8) were subjected to contextual and cued fear conditioning tests. Values that are significantly ...

Additional behavioral tests were performed with wild-type and hsp110/ mice at 18 months of age. These tests included limb-clasping reflexes and the ability to grasp a bar. In terms of limb clasping, 15 to 20% of the hsp110/ mice show some limb clasping at this age. In the case of the ability of hsp110/ mice to grasp a bar, we detected a deficiency in hsp110/ mice performing this test compared to wild-type mice (Fig. (Fig.3D3D).

Hsp110 is present in tau immunocomplexes, and hsp110−/− brain extracts exhibit reduced PPIase activity.

Tau has previously been shown to interact with Pin1 isomerase (43). In addition, Pin1 appears to facilitate cis-trans isomerization of tau at threonine 231 (43, 66), and pin1−/− mice exhibit p-tau, NFTs, and neurodegeneration (42). To determine whether Hsp110 interacts with tau or Pin1, brain extracts from wild-type or hsp110/ mice were subjected to immunoprecipitation analyses using antibody to total tau, and immunoblotting was performed to detect Hsp110 or Pin1 (Fig. (Fig.4A,4A, left immunoblot). The results show that antibody to total tau can immunoprecipitate Hsp110 from wild-type brain extracts (Fig. (Fig.4A,4A, left immunoblot, lane 2), and not from hsp110/ brain extracts as expected (Fig. (Fig.4A,4A, lane 5). We also tested whether total tau can pull down Pin1 as has previously been shown by others (44). We found that antibody to total tau immunoprecipitated Pin1 from both wild-type and hsp110/ brain extracts (Fig. (Fig.4A,4A, left immunoblot, lanes 2 and 5). Expression of tau and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) are presented as controls for immunoprecipitation. In a comparable experiment, brain extracts from wild-type or hsp110/ mice were subjected to immunoprecipitation using antibody to Hsp110 (Fig. (Fig.4A,4A, right immunoblot). Immunoblotting was performed to detect Pin1. The results show that antibody to Hsp110 could immunoprecipitate Pin1 from wild-type brain extracts (Fig. (Fig.4,4, right immunoblot, lane 2). Expression of immunoprecipitated Hsp110 and GAPDH are presented as controls. In a separate experiment, we could also immunoprecipitate Hsp110 and detect tau by immunoblotting (data not shown).

FIG. 4.
Hsp110 interacts with tau and Pin1. (A) Portions (1-mg portions) of brain extracts from 7-month-old wild-type (WT) or hsp110/ mice were subjected to immunoprecipitation (IP) using antibody (Ab) to total tau followed by immunoblotting ...

As the data in Fig. Fig.4A4A indicate, Hsp110 is present in tau immunocomplexes; however, it is not clear whether the interaction of Hsp110 and tau is direct. To determine whether Hsp110 interacts with tau directly, we subjected purified wild-type tau441 and Hsp110 to immunoprecipitation analyses. Immunoblotting of the pulled-down materials suggest that Hsp110 and tau interact directly (Fig. (Fig.4B).4B). These results indicate that Hsp110 interacts with tau and Pin1 and is directly involved in the regulation of the phosphorylation state of tau.

Pin1 activity is critical for proper dephosphorylation of phosphorylated tau (42). Since we found that Pin1 interacts with both tau and Hsp110 in brain extracts, we compared the activity of PPIases in brain extracts from hsp110/ mice to that in wild-type mice (67). As the data in Fig. Fig.4C4C indicate, PPIase activity is significantly lower in brain extracts of hsp110/ mice at 2 or 7 months of age compared to wild-type mice of the same ages. Immunoblotting analyses did not show a significant reduction in the level of Pin1 in the soluble fraction of brain extracts from 2- or 7-month-old wild-type or hsp110/ mice (data not shown).

Hsp70i-deficient mice exhibit age-dependent accumulation of p-tau.

As mentioned before, Hsp110 has been shown to be the NEF for the stress-inducible Hsp70.1/70.3 (named Hsp70i) and the constitutively expressed Hsc70 (14, 56, 62). The activities of these proteins are required for proper protein folding (4, 7). In order to determine whether the Hsp110 partner, Hsp70, is also required for maintenance of proper phosphorylation of tau, we analyzed the appearance of p-tau in the brains of hsp70i/ mice in different age groups. The data show that brain tissue sections of hsp70i/ mice exhibit positive immunoreactivity to p-tau using CP13 or PHF1 antibody (Fig. (Fig.5A)5A) (see Fig. S2 in the supplemental material) and that this increased with age. There was positive immunoreactivity to p-tau at T231 and S262 as well (Fig. S2); however, there was no immunoreactivity using MC1 antibody in the cortex (Fig. S2) or in the forebrain as presented in Fig. Fig.22 for the hsp110/ mice. In addition, positive-staining cells expressing NFTs were present in brain tissue sections of hsp70i−/− mice at 32 weeks of age (Fig. (Fig.5B,5B, panels b to d) but were not present in the wild-type mice (Fig. (Fig.5B,5B, panel a) using Bielschowsky staining. Thioflavin T staining of the cortex (hippocampal region) also showed the presence of NFTs (Fig. (Fig.5C,5C, bottom panel). The number of cells showing NFTs using Bielschowsky staining is presented in the graph in Fig. Fig.5C5C.

FIG. 5.
hsp70i/ mice exhibit age-dependent accumulation of p-tau. (A) In the micrographs shown on the left of the panel, the brains of 28-week-old wild-type (WT) or hsp70i/ mice were fixed and processed. Brain tissue sections ...

We also determined PPIase activity (67) in the brain extracts of mice deficient in Hsp70i. The data indicate that the level of PPIase activity was reduced in brain extracts from hsp70i/ mice compared to wild-type mice and that the reduction was comparable to that observed in hsp110/ mice (Fig. (Fig.5D5D).

These data indicate that hsp70i−/− mice also exhibit defects in maintenance of the phosphorylation state of tau and neurodegeneration. In terms of neurodegeneration, examination of 1-year-old male hsp70i/ mice indicated that 25% of these mice exhibited limb-clasping reflexes.

Expression of protein kinases and phosphatases in brain extracts of mice lacking Hsp110 or Hsp70i.

A number of regulatory protein kinases and phosphatases control the phosphorylation state of tau (22, 47, 50, 53, 54, 57). To determine whether the expression of protein kinases such as AKT, GSK3a or GSK3b, or CDK5 were elevated, immunoblotting experiments were performed to detect expression of these proteins in brain extracts of hsp110/ or hsp70i/ mice compared to wild-type mice (Fig. (Fig.6).6). Data indicate that among the protein kinases tested, only the expression of GSK3a and GSK3b phosphorylated at S9/S21 were significantly higher in hsp110/ mice compared to wild-type mice (Fig. (Fig.6,6, graph). The expression level of GSK3a and GSK3b at S9/S21 in brain extracts from hsp70i/ mice was also increased relative to wild-type mice, although not at significant levels. The expression levels of other proteins, including p35 and its cleavage product p25 (neuron-specific activator of CDK5) were not significantly affected in brain extracts from hsp110/ or hsp70i/ mice. The expression level of PP2A or phosphorylated PP2A (p-PP2A) (Tyr307), a protein phosphatase known to be involved in p-tau dephosphorylation (19, 22), was also not significantly affected in brain extracts from hsp110/ or hsp70i/ mice. GSK3a or GSK3b (GSK3a/b) phosphorylated at T216/279 leads to an increase in activity, and phosphorylation of PP2A at Tyr307 is inhibitory to its activity.

FIG. 6.
Expression levels of AKT, GSK3, CDK5, and PP2A in brain extracts from wild-type (+/+), hsp110/ (−/−), and hsp70i/ mice. Soluble brain extracts of 30-week-old mice (3 to 5 mice; males ...

Previous reports indicate that Hsp90 and Hsp70 as well as their cochaperones are in complexes with tau (11). These complexes, together with Chip ubiquitin ligase, are implicated in p-tau refolding and degradation. To determine whether Hsp110 and other Hsps (e.g., Hsp70) are together in tau immunocomplexes, we immunoprecipitated total tau from brain extracts from young and aged wild-type, hsp110/, and hsp70i/ mice and examined the presence of Hsp110, Hsp90, Hsc70, Hsp70, Hsp60, and Hsp40 in tau complexes (Fig. (Fig.7A).7A). The results show that only Hsp60 was not detected in such complexes (and was used as a negative control), while other Hsps were detected in tau immunocomplexes in brain extracts from wild-type, hsp110/, and hsp70i/ mice. As expected, Hsp110 and Hsp70i were absent from the brain extracts of the corresponding knockout mice. Using tau immunocomplexes, we also determined the levels of Pin1, Chip, GSK3b, and PP2A catalytic subunit (PP2A/C) and PP2A regulatory subunit B55-a, which could be pulled down using antibody to tau (11, 22, 42, 44, 55, 66). Interestingly, the amount of immunoprecipitated GSK3b was significantly reduced in the tau immunocomplexes from brain extracts of aged (8-month-old) hsp110/ mice (Fig. (Fig.7B).7B). This is consistent with previous reports indicating that aggregated p-tau may bind GSK3b and exclude it from interacting with tau that is present in the soluble brain fraction (1). GSK3b appears in tau immunocomplexes but was also reduced in soluble brain extracts from hsp70i/ mice than in wild-type mice. Surprisingly, we found a 4-fold increase in the level of PP2A/C that was bound to tau in soluble brain extracts from aged hsp110/ and hsp70i/ mice. There was a 30 to 50% increase in the B55-a subunit of PP2A in brain extracts from aged hsp110/ and hsp70i/ mice that bound to tau as well (Fig. (Fig.7B).7B). GAPDH was absent in all tau immunoprecipitated groups, and the absence of this protein was used as a control. The amount of pulled-down tau is also presented.

FIG. 7.
Elevated levels of PP2A remain bound to tau in aging hsp110/ and hsp70i/ mice. (A and B) Tau is present in immunocomplexes with specific Hsps, GSK3, PP2A, and Pin1. Brain lysates (1-mg portions) from the indicated groups ...

These results indicate that Hsp110 is associated with tau (or p-tau) together with other Hsps, Chip, Pin1, PP2A, and GSK3b (with the exception of brain extracts from aged hsp110/ mice) in the brain extracts from wild-type, hsp110/ or hsp70i/ mice. The presence of Hsc70 in tau immunocomplexes may compensate for its deficiency in hsp70i/ brain extracts, since some GSK3b remained bound to tau complexes (although at reduced levels compared to wild-type mice) in hsp70i/ aged mice.

From the results presented in Fig. Fig.44 and and5,5, it appears that expression of Hsp110 and Hsp70i is required for proper PPIase activities, which has been shown to be essential for the chaperone function of Hsps as well as reduced p-tau and NFTs (21, 36, 42). In addition, a significant increase in the level of PP2A/C was observed in the soluble brain extracts of aged hsp110/ and hsp70i/ mice, which remained bound to tau. In order to understand the underlying reason for increase in PP2A levels binding to tau, we hypothesized that PP2A remained bound to tau because it may require Hsp110 or Hsp70i for its activity perhaps before its release from the complex. We therefore determined PP2A activity that remained bound to tau. The data presented in Fig. Fig.7C7C indicate that PP2A in tau immunocomplexes of young and aged hsp110/ and hsp70i/ mice exhibited significant reduction in activity compared to PP2A in wild-type mice. These results indicate that PP2A dephosphorylation of tau and its subsequent release from tau require the presence of Hsp110 and Hsp70i.

Accelerated pathology in hsp110−/− Tg2576+ mice compared to Tg2576+ mice.

Hsp70 has been shown to be involved in APP processing in cultured cells, since the presence of Hsp70 lowers the toxicity of β-amyloid (Aβ) derived from APP in cells (39). Hsp70 also protects cultured cells against the toxic effects of soluble Aβ42, a critical contributor to the formation of plaques in AD (39). Indeed, neuropathological hallmarks of AD are NFTs that contain both p-tau and Aβ (3). However, the in vivo evidence that a molecular chaperone may impact APP processing and Aβ accumulation is not known. To determine whether hsp110/ mice exhibit accelerated pathology when crossed with mice carrying the Tg(APPSWE)2576Kha transgene (also known as Tg2576) that overexpress familial Alzheimer's disease (FAD) APPK670N/M671L mutation (29), we generated hsp110/ Tg2576+ mice. We first examined whether antibody to human APP can pull down Hsp110 from brain extracts of Tg2576+ mice, and immunoprecipitation experiments indicate that APP can pull down Hsp110 (Fig. (Fig.8A,8A, left panel, lane 2). The right panel of Fig. Fig.8A8A shows the levels of APP expression in brain extracts of wild-type, Tg2576+, and hsp110/ Tg2576+ mice. Since we observed that hsp110/ brain extracts possessed elevated levels of GSK3a/b S9/S21 and increased GSK3 has also been shown to be an important factor in the formation of NFTs and amyloid plaques in AD (18, 53, 58), we performed immunoprecipitation to detect GSK3b (and GSK3a/b S9/S21) in APP complexes. The results presented in Fig. Fig.8A8A indicate that APP complexes contain abundant levels of GSK3b. However, phosphorylated GSK3 could not be detected, perhaps due to the low level of these species in the brain extracts as evidenced by immunoblotting presented in Fig. Fig.8B.8B. There was a low but comparable level of PP2A/C present in APP immunocomplexes of both groups as well (Fig. (Fig.8A).8A). The absence of GAPDH in APP immunoprecipitation materials is presented as a control. We also performed immunoblotting to detect the level of expression of GSK3a/b, CDK5, p35/p25, and PP2A in brain extracts from Tg2576+ and hsp110/ Tg2576+ mice. The data indicate that the levels of phosphorylated GSK3a/b were 2-fold higher in brain extracts from hsp110/ Tg2576+ mice than in Tg2576+ mice; however, the levels of CDK5, p35/p25, and PP2A were not significantly affected in the brains of these mice.

FIG. 8.
Hsp110 interacts with APP and is involved in APP processing. (A) (Left) Brain extracts from Tg2576+ mice (12-month-old male mice) or hsp110/ Tg2576+ mice (7-month-old male mice) (n = 3) were subjected to immunoprecipitation ...

Since aged hsp110/ mice exhibited NFTs, we stained brain tissue sections of Tg2576+ or hsp110/ Tg2576+ mice for the presence of NFTs. Tg2576+ mice have not been shown to accumulate NFTs, and our data support that (Fig. (Fig.8C);8C); however, 6-month-old hsp110/ Tg2576+ mice did show an abundant number of cells containing NFTs using both Bielschowsky staining (Fig. (Fig.8C,8C, top panels) and thioflavin T staining (Fig. (Fig.8C,8C, bottom panels). The number of cells exhibiting NFTs following Bielschowsky staining was higher in hsp110/ Tg2576+ mice than in hsp110/ mice (Fig. (Fig.8C,8C, graph). The Tg2576+ mice exhibit senile plaques when the mice were 12 months of age and older (29). We therefore examined whether the absence of the hsp110 gene leads to the appearance of senile plaques in hsp110/ Tg2576+ mice at a younger age. Analyses of brain tissue sections from Tg2576+ and hsp110/ Tg2576+ mice indicate that Tg2576+ mice express senile plaques only when the mice were 12 months of age, while the sections from hsp110/ Tg2576+ mice express senile plaques when the mice were 7 months of age using Congo red staining or immunostaining using antibody to Aβ (Fig. (Fig.8D).8D). No plaques were detected in 12-month-old wild-type, hsp110/, or 8-month-old Tg2576+ mice (Fig. (Fig.8D,8D, top panels). Quantitation of Aβ plaques in the mouse lines is presented in the graph in Fig. Fig.8D.8D. Unfortunately, many hsp110/ Tg2576+ mice often died before reaching 7 months of age. The reason for this mortality is not known, but it is most likely due to the lack of Hsp110 and increased pathology in these mice. This prevented extensive analyses of the hsp110/ Tg2576+ mouse line.

The above data indicate that loss of the hsp110 gene leads to early Aβ production and accelerates neurodegeneration in Tg2576+ mice, and lack of Hsp110 plays a critical role in the progression of AD.

α-Secretase and β-secretase activities and presence of Aβ40 and Aβ42 in brain extracts of hsp110−/− and hsp70i−/− mice.

APP is processed by nonamyloidogenic α-secretase at the plasma membrane (75). APP is processed by β- and γ-secretases at endosomes and other sites (46, 51, 75). In order to determine whether brain tissue in wild-type, hsp110/, or hsp70i/ mice possess comparable levels of α- and β-secretase activities, brain extracts from 7-month-old wild-type, hsp110/, and hsp70i/ mice were used to determine the activities of these enzymes. The data presented in Fig. Fig.9A9A indicate that the activity of α-secretase was comparable in mice of all genotypes. The β-secretase activities in brain extracts from wild-type and hsp110/ mice were also not significantly different; however, β-secretase activity in brain extracts from hsp70i/ mice was significantly higher than in the wild-type mice (P < 0.01).

FIG. 9.
hsp110/ mice exhibit wild-type levels of α-secretase and β-secretase activities but show an accelerated increase in the insoluble Aβ42 generation. (A) Soluble brain extracts of 7-month-old male mice (n = ...

We also determined the levels of mouse Aβ40 and Aβ42 in the soluble and insoluble fractions of brain extracts of wild-type, hsp110/, and hsp70i/ mice. The data indicate that the levels of Aβ40 in the soluble fraction, Aβ40 in the insoluble fraction, and Aβ42 in the soluble fraction of brain extracts were comparable in mice of all genotypes (Fig. (Fig.9B).9B). However, Aβ42 levels in the insoluble fraction of brain extracts from hsp110/ and hsp70i/ mice were significantly different than in the wild-type mice (Fig. (Fig.9B9B).

Tg2576+ mice express a mutant form of APP. Therefore, we then asked whether brain tissue extracts derived from hsp110/ Tg2576+ mice express more Aβ40 or the toxic species Aβ42 in the soluble or insoluble brain extracts compared to Tg2576+ mice. The data presented in Fig. Fig.9C9C indicate that soluble Aβ40 and Aβ42 were not significantly different in 7-month-old hsp110+/ Tg2576+ and hsp110/ Tg2576+ mice. However, Aβ40 and Aβ42 levels in the soluble brain extracts in Tg2576+ mice were significantly different than the levels in the wild-type mice (Fig. (Fig.9C).9C). We detected significantly higher levels of both Aβ40 and Aβ42 in insoluble brain extracts from 7-month-old hsp110/ Tg2576+ mice than in hsp110+/ Tg2576+ littermates (Fig. (Fig.9D,9D, right two panels). For comparison, 1-year-old Tg2576+ mice were also found to possess significant amounts of Aβ40 and Aβ42 in the insoluble brain extracts, as expected (Fig. (Fig.9D,9D, right panels).

Indeed, immunoblotting experiments showed that soluble brain extracts of hsp110/ Tg2576+ mice contained the sAPPβsw mutant form as early as 6 weeks of age (Fig. (Fig.9E),9E), while hsp110+/ Tg2576+ mice did not express this fragment. As expected, both Tg2576+ mice tested at 10 months of age and hsp110/ Tg2576+ mice tested at 6 months of age expressed this APP species. Taken together, these data strongly suggest that Hsp110 is critical for delaying insoluble Aβ42 accumulation in this mouse model of AD.

Hsp110 is expressed in brain tissue from patients with Alzheimer's disease and healthy humans.

Hsp90 and Hsp70 have been shown to colocalize with senile plaques in the brains of patients with Alzheimer's disease (23, 73). As noted before, it has been shown that Hsp110 is expressed in neurons; however, whether it colocalizes with Aβ in senile plaques in the brains of AD patients has not been investigated. To determine the expression of Hsp110 in human brain tissue, paraffin-embedded sections of healthy human brain and Alzheimer's brain tissue samples were immunostained using antibody to Hsp110 or using antibodies to Hsp110 and Aβ (Fig. 10A and B). The data indicate that Hsp110 is expressed in brain tissue sections from patients with Alzheimer's disease in close proximity to Aβ plaques (Fig. 10A). Hsp110 immunostaining appears in close proximity to Aβ (Fig. 10A, arrows) in cells that have intact nuclei (stained with DAPI) (top panel, arrowheads) or in those cells with no obvious nuclear staining (perhaps apoptotic cells or protein aggregates) but in the proximity of Aβ deposits (Fig. 10A, bottom panel, arrowheads).

FIG. 10.
Levels of expression of Hsp110 and Aβ in brain tissue samples from healthy individuals and patients with Alzheimer's disease. (A) Paraffin-embedded human tissue samples of brain sections from patients with Alzheimer's disease (Biochain) (hippocampal ...


Accumulating evidence indicates a role for molecular chaperones in protein misfolding diseases (4, 49). Brain tissue sections from AD patients as well as other tauopathies, such as Pick's disease, corticobasal degeneration, progressive supranuclear palsy, and frontotemporal dementia and parkinsonism linked to chromosome 17 (FTDP-17), have been shown to be immunopositive for molecular chaperones (Hsp90, Hsp70, and their cochaperones) as well as the Chip ubiquitin ligase (25, 38, 59). Furthermore, Hsp70 (and Hsc70) have been shown to associate with hyperphosphorylated tau (p-tau), and together with Chip ubiquitin ligase, they regulate folding, ubiquitination, and degradation of tau and p-tau (4, 11-13). Enhanced expression of Hsp70 decreases tau aggregation and increases tau degradation, indicating that Hsp70 may promote tau degradation (52). Chip ubiquitin ligase activity is dependent on the Hsp70 or Hsp90 molecular chaperone (10, 79). Chip/ brain tissue accumulates p-tau, but these mice do not exhibit NFTs (11). Others have shown that reduction in Hsp90 using EC102, an inhibitor of ATPase activity of Hsp90, decreases tau levels both in vitro and in vivo. Inhibition of Hsp90 using small interfering RNA (siRNA) leads to inhibition of the EC102 effects on tau (11). These data place Hsp90 as part of the tau refolding and degradation pathway (11). Accumulating data have led to a model suggesting that p-tau binds to Chip and Hsp70 and is subsequently passed on to Hsp90 and its cochaperones for further folding and/or degradation (11, 21, 65). Other components of the tau phosphorylation, folding, and degradation machinery are the peptidyl prolyl isomerases (9, 36, 42). The Hsp90 cochaperone, FK506-binding protein 51 (FKBP-51), that binds Hsp90 and possesses PPIase activity regulates phosphorylation and clearance of tau (36). FKBP-52 also binds to p-tau, and its overexpression in PC12 cells inhibits accumulation of tau, reducing neurite growth (9). Perhaps more-direct in vivo evidence confirming a role for PPIases in the phosphorylation state of tau is data showing that deletion of Pin1 isomerase in mice leads to an age-dependent accumulation of p-tau and NFTs and neurodegeneration (42, 44, 51).

Tau is a microtubule-associated protein that assists microtubule assembly, sustains neuronal integrity, and is involved in axonal transport (3, 38). Tau is predominantly expressed in neurons/axons and is phosphorylated on many serine and threonine residues. Under physiological conditions, tau is phosphorylated on 2 or 3 residues (20). Tau phosphorylation and dephosphorylation occurs by a number of protein kinases, such as GSK3, CDK5, AKT, and phosphoprotein phosphatases, such as PP2A, which regulates tau association and dissociation with the microtubules (3, 6, 20, 22, 34). Interestingly, dephosphorylation of substrates by PP2A is also regulated by a new class of PPIases known as PTPAs (phosphatase two A phosphatase activators), indicating that PPIases are an important regulator of protein homeostasis. Indeed, PPIases such as PTPAs are in a different class from Pin1 isomerase (37). While PTPA is a known activator of PP2A, Pin1 affects phosphorylation of a number of substrates, including isomerization of Thr231 of tau and Thr668 of APP (42, 44, 51). In the disease state, tau remains phosphorylated and forms aggregates (3). Aggregated tau forms NFTs, leading to neuronal death. Although the regulation of tau phosphorylation, dephosphorylation, and degradation has not been fully revealed, the data indicate that accumulation of phosphorylated tau and its aggregation will lead to neurodegeneration (3, 5, 38, 69).

Further investigation of the roles of molecular chaperones in tau homeostasis demonstrates that Hsp110, a NEF for Hsp70, interacts with tau (and p-tau), Pin1, and PP2A and that brain tissue from hsp110/ (and hsp70i/) mice accumulates p-tau and NFTs. With the exception of GSK3a/b S9/S21, which showed increased phosphorylation in brain tissue from hsp110/ mice, the levels of other protein kinases tested in brain tissue from hsp110/ or hsp70i/ mice were comparable to that in wild-type mice. Although GSK3a/b p-S9/S21 (phosphorylated S9/S21) is known to reduce its activity in some cells, this phospho-GSK3 has been shown to phosphorylate tau in neurons (24, 80). However, GSK3 is entirely eliminated from the tau immunocomplexes in the absence of Hsp110 in aged mice. Upon lysis of Sarkosyl-insoluble brain extracts from aged hsp110/ mice in higher SDS concentrations (0.3%), we could pull down 14-3-3zeta and GSK3b together with p-tau using PHF1 antibody (data not shown) as has previously been reported (1). These data suggest that the complexes containing 14-3-3zeta, GSK3b, and p-tau form aggregates and are present in the insoluble fractions of brain lysates from hsp110/ mice. However, perhaps more revealing, we found a significant increase in PP2A association with tau in soluble brain extracts from aged hsp110/ and hsp70i/ mice but with dramatically reduced activity. These data strongly indicate a role for the Hsp110/Hsp70 molecular machine in the activation of PP2A, and perhaps facilitation of its release from tau. The mechanism underlying lower PP2A expression or activity in the brains of AD patients is not known, but it could be due to higher levels of PP2A inhibitors, B55 regulatory subunit, inhibitory Tyr307 phosphorylation, Leu309 demethylation, or a decrease in PP2A/C subunit levels (22, 37, 43). We did not find a significant reduction in the level of PP2A/C or PP2A-B55a or in the level of PP2A Tyr307 phosphorylation in brain tissue extracts from hsp110/ or hsp70/ mice. As noted before, protein phosphatase 2A protein activator PTPA is a peptidyl prolyl cis-trans isomerase that activates PP2A (37). We found a significant reduction in the activities of both PPIases and PP2A (bound to tau) in the soluble brain extracts in hsp110/ and hsp70i/ mice. The general substrate that we selected to determine PPIase activities in brain extracts from wild-type and knockout mice may not include the activity of all the PPIases that have been shown to affect tau phosphorylation state; however, lower activities of specific PPIases (including Pin1 or PTPA) as well as PP2A in brain extracts of hsp110/ and hsp70i/ mice are the likely causes for the observed phenotype.

As noted before, the activities of PP2A and isomerases are critical in AD pathology (9, 19, 26, 36, 42, 43, 51, 72). In terms of isomerases, Pin1 has been shown to bind the phosphorylated Thr668 residue of APP, increasing its isomerization, processing, and Aβ production (51). Indeed, pin1/ mice exhibit increases in APP processing and neurodegeneration (51). Our results implicate the activity of Hsp110 molecular chaperone in processing of APP. We not only demonstrate that Hsp110 interacts with APP, we also show that hsp110/ Tg2576+ mice exhibit Aβ plaques at a younger age than Tg2576+ mice. Normally, in Tg2576+ mice, plaques appear around 1 year of age, while we could detect plaques in hsp110/ Tg2576+ mice as early as 7 months. In addition, hsp110/ Tg2576+ mice exhibit an increase in amyloidogenic APPβsw processing, and indeed, hsp110/ Tg2576+ mice accumulate insoluble Aβ40 and Aβ42 at a significantly accelerated pace (within 6 months of age) compared to hsp110+/ Tg2576+ littermates. hsp110/ and hsp70i/ mice also accumulate insoluble Aβ42 in mice tested at 1 year of age compared to wild-type mice. These data suggest that the absence of the hsp110 gene leads to a defect in APP processing and Aβ accumulation. Interestingly, the activities of both α-secretase and β-secretase in brain extracts from hsp110/ mice are at levels comparable to those in wild-type mice. Additionally, in vitro cell culture studies suggest that Hsp90 and Hsp70, together with Chip ubiquitin ligase, interact and affect the metabolism of βAPP (39). Data suggest an interaction between Chip, Hsps, and Aβ42 and that the presence of these molecules reduces build-up of Aβ42 (39). Comparably, Pin1 overexpression has been shown to decrease Aβ secretion, and reduction in Pin1 increases Aβ secretion (51). Mice deficient in Pin1 overexpressing the APP mutant (Tg2576+) (51) selectively exhibit increases in amyloidogenic processing of APP and generation of insoluble Aβ42. Indeed, Pin1 isomerization of APP controls Aβ production and APP processing (51). Therefore, we envision that nonamyloidogenic APP processing, Aβ production, and APP degradation may require, at least in part (or in specific cellular compartments), Hsp110, Hsp70, and PP2A as well as Pin1, Hsp90, and its cochaperones, and Chip (11, 12, 39, 51). Deletion of either Pin1 (51) or Hsp110 (presented here) in vivo leads to amyloidogenic processing of APP and toxic Aβ42 accumulation. Although we did not find significant changes in the levels of PP2A in brain extracts of hsp110/ mice (with or without the presence of the Tg2576+ transgene) or significant increase in PP2A binding to APPβsw in the absence of Hsp110, lower levels of PP2A activity that likely are present in the brain extracts of hsp110/ mice may affect the observed alterations in the APP processing that have previously been shown by others (22, 43).

In conclusion, we have provided genetic evidence that lack of hsp110 (or Hsp70) in mice is associated with tauopathy and neurodegeneration, suggesting that molecular chaperones are critical in neurodegenerative diseases, such as AD and other tauopathies. Hsp110-deficient mice may provide a good model system to define mechanisms involved in tau hyperphosphorylation, APP processing, and Aβ plaque accumulation.

Supplementary Material

[Supplemental material]


We thank Lei Huang (Medical College of Georgia) for advice on experimental procedures. The antibodies to total tau and phosphorylated tau were generous gifts from P. Davies (Albert Einstein College of Medicine) and P. Seubert (Elan Pharmaceuticals).

This work was supported by NIH grants CA062130, CA132640, and VA award 1I01BX000161 (N.F.M.) and by NIH grant CA121951 (D.M.).


[down-pointing small open triangle]Published ahead of print on 2 August 2010.

Supplemental material for this article may be found at http://mcb.asm.org/.


1. Agarwal-Mawal, A., H. Y. Qureshi, P. W. Cafferty, Z. Yuan, D. Han, R. Lin, and K. H. Paudel. 2003. 14-3-3 connects glycogen synthase kinase-3β to tau within a brain microtubule-associated tau phosphorylation complex. J. Biol. Chem. 278:12722-127728. [PubMed]
2. Andreasson, C., J. Fiaux, H. Rampelt, M. P. Mayer, and B. Bukau. 2008. Hsp110 is a nucleotide-activated exchange factor for Hsp70. J. Biol. Chem. 238:8877-8884. [PubMed]
3. Ballatore, C., V. M. Lee, and J. Q. Trojanowski. 2007. Tau-mediated neurodegeneration in Alzheimer's disease and related disorders. Nat. Rev. Neurosci. 8:663-672. [PubMed]
4. Barral, J. M., S. A. Broadley, G. Schaffar, and F. U. Hartl. 2004. Roles of molecular chaperones in protein misfolding diseases. Semin. Cell Dev. Biol. 15:17-29. [PubMed]
5. Billingsley, M. L., and R. L. Kincaid. 1997. Regulated phosphorylation and dephosphorylation of tau protein: effects on microtubule interaction, intracellular trafficking and neurodegeneration. Biochem. J. 323:577-591. [PMC free article] [PubMed]
6. Buee, L., T. Bussiere, V. Buee-Scherrer, A. Delacourte, and P. R. Hof. 2000. Tau protein isoforms, phosphorylation and role in neurodegenerative disorders. Brain Res. Brain Res. Rev. 33:95-130. [PubMed]
7. Bukau, B., J. Weissman, and A. Horwich. 2006. Molecular chaperones and protein quality control. Cell 125:443-451. [PubMed]
8. Cairns, N. J., V. M.-Y. Lee, and J. Q. Trojanowski. 2004. The cytoskeleton in neurodegenerative diseases. J. Pathol. 204:438-449. [PMC free article] [PubMed]
9. Chambraud, B., E. Sardin, J. Giustiniani, O. Dounane, M. Schumacher, M. Goedert, and E. E. Baulieu. 2010. A role for FKBP52 in Tau protein function. Proc. Natl. Acad. Sci. U. S. A. 107:2658-2663. [PMC free article] [PubMed]
10. Connell, P., C. A. Ballinger, J. Jiang, Y. Wu, L. J. Thompson, J. Hohfeld, and C. Patterson. 2001. The co-chaperone CHIP regulates protein triage decisions mediated by heat shock proteins. Nat. Cell Biol. 3:93-96. [PubMed]
11. Dickey, C. A., A. Kamal, K. Lundgren, N. Klosak, R. M. Bailey, J. Dunmore, P. Ash, S. Shoraka, J. Zlatkovic, C. B. Eckman, C. Patterson, D. W. Dickson, N. S. Nahman, M. Hutton, F. Burrows, and L. Petrucelli. 2007. The high-affinity HSP90-CHIP complex recognizes and selectively degrades phosphorylated tau client proteins. J. Clin. Invest. 117:648-658. [PMC free article] [PubMed]
12. Dickey, C. A., M. Yue, W. L. Lin, D. W. Dickson, J. H. Dunmore, W. C. Lee, C. Zehr, G. West, S. Cao, A. M. Clark, G. A. Caldwell, K. A. Caldwell, C. Eckman, C. Patterson, M. Hutton, and L. Petrucelli. 2006. Deletion of the ubiquitin ligase CHIP leads to the accumulation, but not the aggregation, of both endogenous phospho- and caspase-3-cleaved tau species. J. Neurosci. 26:6985-6996. [PubMed]
13. Dou, F., W. J. Netzer, K. Tanemura, F. Li, F. U. Hartl, A. Takashima, G. K. Gouras, P. Greengard, and H. Xu. 2003. Chaperones increase association of tau protein with microtubules. Proc. Natl. Acad. Sci. U. S. A. 100:721-726. [PMC free article] [PubMed]
14. Dragovic, Z., S. A. Broadley, Y. Shomura, A. Bracher, and F. U. Hartl. 2006. Molecular chaperones of the Hsp110 family act as nucleotide exchange factors of Hsp70s. EMBO J. 25:2519-2528. [PMC free article] [PubMed]
15. Easton, D. P., Y. Kaneko, and J. R. Subjeck. 2000. The hsp110 and Grp1 70 stress proteins: newly recognized relatives of the Hsp70s. Cell Stress Chaperones 5:276-290. [PMC free article] [PubMed]
16. Esser, C., S. Alberti, and J. Höhfeld. 2004. Cooperation of molecular chaperones with the ubiquitin/proteasome system. Biochim. Biophys. Acta 1695:171-188. [PubMed]
17. Gashegu, J., R. Ladha, N. Vanmuylder, C. Philippson, F. Bremer, M. Rooze, and S. Louryan. 2007. HSP110, caspase-3 and -9 expression in physiological apoptosis and apoptosis induced by in vivo embryonic exposition to all-trans retinoic acid or irradiation during early mouse eye development. J. Anat. 210:532-541. [PMC free article] [PubMed]
18. Giese, K. P. 2009. GSK-3: a key player in neurodegeneration and memory. IUBMB Life 61:516-521. [PubMed]
19. Gong, C. X., I. Grungke-Iqbal, and K. Iqbal. 1994. Dephosphorylation of Alzheimer's disease abnormally phosphorylated tau by protein phosphatase 2A. Neuroscience 61:765-772. [PubMed]
20. Gong, C. X., F. Liu, F. I. Grundke-Iqbal, and K. Iqbal. 2005. Post-translational modifications of tau protein in Alzheimer's disease. J. Neural Transm. 112:813-838. [PubMed]
21. Goryunov, D., and R. K. H. Liem. 2007. CHIP-ping away at tau. J. Clin. Invest. 17:590-592. [PMC free article] [PubMed]
22. Götz, J., A. Gladbach, L. Pennanen, J. van Eersel, A. Schild, D. David, and L. M. Ittner. 12 September 2009. Animal models reveal role for tau phosphorylation in human disease. Biochim. Biophys. Acta doi:. [Epub ahead of print.]10.1016/j.bbadis.2009.09.008 [PubMed] [Cross Ref]
23. Guerreiro, N., M. Staufenbiel, and B. Gomez-Mancilla. 2008. Proteomic 2-D DIGE profiling of APP23 transgenic mice brain from pre-plaque and plaque phenotypes. J. Alzheimers Dis. 13:17-30. [PubMed]
24. Hashiguchi, M., K. Sobue, and H. K. Pudel. 2000. 14-3-3zeta is an effector of tau protein phosphorylation. J. Biol. Chem. 275:25247-25254. [PubMed]
25. Hatakeyama, S., M. Matsumoto, T. Kamura, M. Murayama, D. H. Chui, E. Planel, R. Takahashi, K. I. Nakayama, and A. Takashima. 2004. U-box protein carboxyl terminus of Hsc70-interacting protein (CHIP) mediates poly-ubiquitylation preferentially on four-repeat Tau and is involved in neurodegeneration of tauopathy. J. Neurochem. 91:299-307. [PubMed]
26. Holzer, M., M. K. Bruckner, M. Beck, V. Bigl, and T. Arendt. 2000. Modulation of APP processing and secretion by okadaic acid in primary guinea pig neurons. J. Neural Transm. 107:451-461. [PubMed]
27. Homma, S., X. Jin, G. Wang, N. Tu, J. Min, N. Yanasak, and N. F. Mivechi. 2007. Demyelination, astrogliosis, and accumulation of ubiquitinated proteins, hallmarks of CNS disease in hsf1-deficient mice. J. Neurosc. 27:7974-7986. [PubMed]
28. Hrizo, S. L., V. Gusarova, D. M. Habiel, J. L. Goeckeler, E. A. Fisher, and J. L. Brodsky. 2007. The Hsp110 molecular chaperone stabilizes apolipoprotein B from endoplasmic reticulum-associated degradation (ERAD). J. Biol. Chem. 282:32665-32675. [PMC free article] [PubMed]
29. Hsiao, K., P. Chapman, S. Nilsen, C. Eckman, Y. Harigaya, S. Younkin, F. Yang, and G. Cole. 1996. Correlative memory deficits, Abeta elevation, and amyloid plaques in transgenic mice. Science 274:99-102. [PubMed]
30. Hu, Y., and N. F. Mivechi. 2006. Association and regulation of heat shock transcription factor 4b with both extracellular signal-regulated kinase mitogen-activated protein kinase and dual-specificity tyrosine phosphatase DUSP26. Mol. Cell. Biol. 26:3282-3294. [PMC free article] [PubMed]
31. Huang, L., J. Min, S. Maters, N. F. Mivechi, and D. Moskophidis. 2007. Insights into the function and regulation and of small hsp25 (HSPB1) in mouse model with targeted gene disruption. Genesis 45:487-501. [PubMed]
32. Huang, L., N. F. Mivechi, and D. Moskophidis. 2001. Insights into regulation and function of the major stress-induced hsp70 molecular chaperone in vivo: analysis of mice with targeted gene disruption of the hsp70.1 or hsp70.3 genes. Mol. Cell. Biol. 21:8575-8591. [PMC free article] [PubMed]
33. Hylander, B. L., X. Chen, P. C. Graf, and J. R. Subjeck. 2000. The distribution and localization of hsp110 in brain. Brain Res. 869:49-55. [PubMed]
34. Jellinger, K. A. 2003. General aspects of neurodegeneration. J. Neural Transm. Suppl. 65:101-144. [PubMed]
35. Jicha, G. A., R. Bowser, I. G. Kazam, and P. Davies. 1997. Alz-50 and MC-1, a new monoclonal antibody raised to paired helical filaments, recognize conformational epitopes on recombinant tau. J. Neurosci. Res. 48:128-132. [PubMed]
36. Jinwal, U. K., J. Koren III, S. I. Borysov, A. B. Schmid, J. F. Abisambra, L. J. Blair, A. G. Johnson, J. R. Jones, C. L. Shults, J. C. O'Leary III, Y. Jin, J. Buchner, M. B. Cox, and C. A. Dickey. 2010. The Hsp90 cochaperone, FKBP51, increases Tau stability and polymerizes microtubules. J. Neurosci. 30:591-599. [PMC free article] [PubMed]
37. Jordens, J., V. Janssens, S. Longin, I. Stevens, E. Martens, G. Bultynck, Y. Engelborghs, E. Lescrinier, E. Waelkens, J. Goris, and C. Van Hoof. 2006. The protein phosphatase 2A phosphatase activator is a novel peptidyl-prolyl cis/trans-isomerase. J. Biol. Chem. 281:6349-6357. [PubMed]
38. Kosik, K. S., and H. Shimura. 2005. Phosphorylated tau and the neurodegenerative foldopathies. Biochim. Biophys. Acta 1739:298-310. [PubMed]
39. Kumar, P., R. K. Ambasta, V. Veereshwarayya, K. M. Rosen, K. S. Kosik, H. Band, R. Mestril, C. Patterson, and H. W. Querfurth. 2007. CHIP and HSPs interact with beta-APP in a proteasome-dependent manner and influence Abeta metabolism. Hum. Mol. Genet. 16:848-864. [PubMed]
40. Ladurelle, N., B. Eychenne, D. Denton, J. Blair-West, M. Schumacher, P. Robel, and E. Baulieu. 2000. Prolonged intracerebroventricular infusion of neurosteroids affects cognitive performances in the mouse. Brain Res. 858:371-379. [PubMed]
41. Liang, P., and T. H. MacRae. 1997. Molecular chaperones and the cytoskeleton. J. Cell Sci. 110:1431-1440. [PubMed]
42. Liou, Y. C., A. Sun, X. Z. Ryo, Z. X. Zhou, H. K. Yu, T. Huang, R. Uchida, G. Bronson, X. Bing, T. Li, T. Hunter, and K. P. Lu. 2003. Role of the prolyl isomerase Pin1 in protecting against age-dependent neurodegeneration. Nature 424:556-561. [PubMed]
43. Liu, R., and J.-Z. Wang. 2009. Protein phosphatase 2A in Alzheimer's disease. Pathophysiology 16:273-277. [PubMed]
44. Lu, P. J., G. Wulf, X. Z. Zhou, P. Davies, and K. P. Lu. 1999. The prolyl isomerase Pin1 restores the function of Alzheimer-associated phosphorylated tau protein. Nature 399:784-788. [PubMed]
45. Ma, M. X., Y. M. Chen, J. He, T. Zeng, and J. H. Wang. 2007. Effects of morphine and its withdrawal on Y-maze spatial recognition memory in mice. Neuroscience 147:1059-1065. [PubMed]
46. Mattson, M. P. 2004. Pathways towards and away from Alzheimer's disease. Nature 430:631-639. [PMC free article] [PubMed]
47. Michel, G., M. Mercken, M. Murayama, K. Noguchi, K. Ishiguro, K. Imahori, and A. Takashima. 1998. Characterization of tau phosphorylation in glycogen synthase kinase-3beta and cyclin dependent kinase-5 activator (p23) transfected cells. Biochim. Biophys. Acta 1380:177-182. [PubMed]
48. Min, J., Y. Zhang, D. Moskophidis, and N. F. Mivechi. 2004. Unique contribution of heat shock transcription factor 4 in ocular lens development and fiber cell differentiation. Genesis 40:205-217. [PubMed]
49. Muchowski, P. J., and J. L. Wacker. 2005. Modulation of neurodegeneration by molecular chaperones. Nat. Rev. Neurosci. 6:11-22. [PubMed]
50. Noble, W., V. Olm, K. Takata, E. Casey, O. Mary, J. Meyerson, K. Gaynor, J. LaFrancois, L. Wang, T. Kondo, P. Davies, M. Burns, Veeranna, R. Nixon, D. Dickson, Y. Matsuoka, M. Ahlijanian, L. F. Lau, and K. Duff. 2003. Cdk5 is a key factor in tau aggregation and tangle formation in vivo. Neuron 38:555-565. [PubMed]
51. Pastorino, L., A. Sun, P. J. Lu, X. Z. Zhou, M. Balastik, G. Finn, G. Wulf, J. Lim, S. H. Li, X. Li, W. Xia, L. K. Nicholson, and K. P. Lu. 2006. The prolyl isomerase Pin1 regulates amyloid precursor protein processing and amyloid-beta production. Nature 440:528-534. [PubMed]
52. Petrucelli, L., D. Dickson, K. Kehoe, J. Taylor, H. Snyder, A. Grover, M. De Lucia, E. McGowan, J. Lewis, G. Prihar, J. Kim, W. H. Dillmann, S. E. Browne, A. Hall, R. Voellmy, Y. Tsuboi, T. M. Dawson, B. Wolozin, J. Hardy, and M. Hutton. 2004. CHIP and Hsp70 regulate tau ubiquitination, degradation and aggregation. Hum. Mol. Genet. 13:703-714. [PubMed]
53. Phiel, C. J., C. A. Wilson, V. M. Lee, and P. S. Klein. 2003. GSK3 alpha regulates production of Alzheimer's disease amyloid beta peptide. Nature 432:435-439. [PubMed]
54. Planel, E., X. Sun, and A. Takashima. 2002. Role of GSK3 beta in Alzheimer's disease pathology. Drug Dev. Res. 56:491-510.
55. Plattner, F., M. Angelo, and K. P. Giese. 2006. The roles of cyclin-dependent kinase 5 and glycogen synthase kinase 3 in tau hyperphosphorylation. J. Biol. Chem. 281:25457-25465. [PubMed]
56. Polier, S., Z. Dragovic, F. U. Hartl, and A. Bracher. 2008. Structural basis for the cooperation of Hsp70 and Hsp110 chaperones in protein folding. Cell 133:1068-1079. [PubMed]
57. Rockenstein, E., M. Torrance, A. Adame, M. Mante, P. Bar-on, J. B. Rose, L. Crews, and E. Masliah. 2007. Neuroprotective effects of regulators of the glycogen synthase kinase-3beta signaling pathway in a transgenic model of Alzheimer's disease are associated with reduced amyloid precursor protein phosphorylation. J. Neurosci. 27:1981-1991. [PubMed]
58. Ryder, J., Y. Su, F. Liu, B. Li, Y. Zhou, and B. Ni. 2003. Divergent roles of GSK3 and CDK5 in APP processing. Biochem. Biophys. Res. Commun. 312:922-929. [PubMed]
59. Sahara, N., S. Maeda, Y. Yoshiike, T. Mizoroki, S. Yamashita, M. Murayama, J. M. Park, Y. Saito, S. Murayama, and A. Takashima. 2007. Molecular chaperone-mediated tau protein metabolism counteracts the formation of granular tau oligomers in human brain. J. Neurosci. Res. 85:3098-3108. [PubMed]
60. Saito, Y., N. Yamagishi, and T. Hatayama. 2007. Different localization of Hsp105 family proteins in mammalian cells. Exp. Cell Res. 313:3707-3717. [PubMed]
61. Saura, C. A., G. Chen, S. Malkani, S. Y. Choi, R. H. Takahashi, D. Zhang, G. K. Gouras, A. Kirkwood, R. G. Morris, and J. Shen. 2005. Conditional inactivation of presenilin 1 prevents amyloid accumulation and temporarily rescues contextual and spatial working memory impairments in amyloid precursor protein transgenic mice. J. Neurosci. 25:6755-6764. [PubMed]
62. Schuermann, J. P., J. Jiang, J. Cuellar, O. Llorca, L. Wang, L. E. Gimenez, S. Jin, A. B. Taylor, B. Demeler, K. A. Morano, P. J. Hart, J. M. Valpuesta, E. M. Lafer, and R. Sousa. 2008. Structure of the Hsp110:Hsc70 nucleotide exchange machine. Mol. Cell 31:232-243. [PMC free article] [PubMed]
63. Sereda, M., I. Griffiths, A. Puhlhofer, H. Stewart, M. J. Rossner, F. Zimmerman, J. P. Magyar, A. Schneider, E. Hund, H. M. Meinck, U. Suter, and K. A. Nave. 1996. A transgenic rat model of Charcot-Marie-Tooth disease. Neuron 1:1049-1060. [PubMed]
64. Shaner, L., R. Sousa, and K. A. Morano. 2006. Characterization of Hsp70 binding and nucleotide exchange by the yeast Hsp110 chaperone Sse1. Biochemistry 45:15075-15084. [PMC free article] [PubMed]
65. Shimura, H., D. Schwartz, S. P. Gygi, and K. S. Kosik. 2004. CHIP-Hsc70 complex ubiquitinates phosphorylated tau and enhances cell survival. J. Biol. Chem. 279:4869-4876. [PubMed]
66. Smet, C., A. V. Sambo, J. M. Wieruszeski, A. Leroy, I. Landrieu, L. Buee, and G. Lippens. 2004. The peptidyl prolyl cis/trans-isomerase Pin1 recognizes the phospho-Thr212-Pro213 site on Tau. Biochemistry 43:2032-2040. [PubMed]
67. Sultana, R., D. Boyd-Kimball, H. F. Poon, J. Cai, W. M. Pierce, J. B. Klein, et al. 2006. Oxidative modification and down-regulation of Pin1 in Alzheimer's disease hippocampus: a redox proteomics analysis. Neurobiol. Aging 27:918-925. [PubMed]
68. Terry, A. V. J., and S. P. Mahadik. 2007. Time-dependent cognitive deficits associated with first and second generation antipsychotics: cholinergic dysregulation as a potential mechanism. J. Pharmacol. Exp. Ther. 320:961-968. [PubMed]
69. Trinczek, B., J. Biernat, K. Baumann, E. M. Mandelkow, and E. Mandelkow. 1995. Domains of tau protein, differential phosphorylation, and dynamic instability of microtubules. Mol. Biol. Cell 6:1887-1902. [PMC free article] [PubMed]
70. Tsapara, A., K. Matter, and M. S. Balda. 2006. The heat-shock protein Apg-2 binds to the tight junction protein ZO-1 and regulates transcriptional activity of ZONAB. Mol. Biol. Cell 17:1322-1330. [PMC free article] [PubMed]
71. Uemura, K., A. Kuzuya, Y. Shimozono, N. Aoyagi, K. Ando, S. Shimohama, and A. Kinoshita. 2007. GSK3beta activity modifies the localization and function of presenilin 1. J. Biol. Chem. 282:15823-15832. [PubMed]
72. Wang, J. Z., I. Grundke-Iqbal, and K. Iqbal. 2007. Kinases and phosphatases and tau sites involved in Alzheimer neurofibrillary degeneration. Eur. J. Neurosci. 25:59-68. [PMC free article] [PubMed]
73. Wilhelmus, M. M., R. M. de Waal, and M. M. Verbeek. 2007. Heat shock proteins and amateur chaperones in amyloid-beta accumulation and clearance in Alzheimer's disease. Mol. Neurobiol. 35:203-216. [PMC free article] [PubMed]
74. Williams, D. R. 2006. Tauopathies: classification and clinical update on neurodegenerative diseases associated with microtubule-associated protein tau. Intern. Med. J. 36:652-660. [PubMed]
75. Wilquet, V., and B. De Strooper. 2004. Amyloid-beta precursor protein processing in neurodegeneration. Curr. Opin. Neurobiol. 14:582-588. [PubMed]
76. Xu, Y., Y. Chen, P. Zhang, P. D. Jeffrey, and Y. Shi. 2008. Structure of a protein phosphatase 2A holoenzyme: insights into B55-mediated tau dephosphorylation. Mol. Cell 31:873-885. [PMC free article] [PubMed]
77. Yamagishi, N., K. Ishihara, Y. Saito, and T. Hatayama. 2006. Hsp105 family proteins suppress staurosporine-induced apoptosis by inhibiting the translocation of Bax to mitochondria in HeLa cells. Exp. Cell Res. 312:3215-3223. [PubMed]
78. Yamashita, H., J. Kawamata, K. Okawa, R. Kanki, T. Nakamizo, T. Hatayama, K. Yamanaka, R. Takahashi, and S. Shimohama. 2007. Heat-shock protein 105 interacts with and suppresses aggregation of mutant Cu/Zn superoxide dismutase: clues to a possible strategy for treating ALS. J. Neurochem. 102:1497-1505. [PubMed]
79. Young, J. C., I. Moarefi, and F. U. Hartl. 2001. Hsp90: a specialized but essential protein-folding tool. J. Cell Biol. 154:267-273. [PMC free article] [PubMed]
80. Yuan, Z., A. Agarwal-Mawal, and H. K. Paudel. 2004. 14-3-3 binds to and mediates phosphorylation of microtubule-associated tau protein by Ser9-phosphorylated glycogen synthase kinase 3 beta in the brain. J. Biol. Chem. 279:26105-26114. [PubMed]
81. Zhang, Y., L. Huang, J. Zhang, D. Moskophidis, and N. F. Mivechi. 2002. Targeted disruption of hsf1 leads to lack of thermotolerance and defines tissue-specific regulation for stress-inducible Hsps. J. Cell Biochem. 86:376-393. [PubMed]

Articles from Molecular and Cellular Biology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...