Denervation-related alterations and biological activity of miRNAs contained in exosomes released by skeletal muscle fibers

Exosomes are vesicles released by many eukaryotic cells; their cargo includes proteins, mRNA and microRNA (miR) that can be transferred to recipient cells and regulate cellular processes in an autocrine or paracrine manner. While cells of the myoblast lineage secrete exosomes, it is not known whether skeletal muscle fibers (myofibers) release exosomes. In this study, we found that cultured myofibers release nanovesicles that have bilamellar membranes and an average size of 60–130 nm, contain typical exosomal proteins and miRNAs and are taken up by C2C12 cells. miR-133a was found to be the most abundant myomiR in these vesicles while miR-720 was most enriched in exosomes compared to parent myofibers. Treatment of NIH 3T3 cells with myofiber-derived exosomes downregulated the miR-133a targets proteins Smarcd1 and Runx2, confirming that these exosomes have biologically relevant effects on recipient cells. Denervation resulted in a marked increase in miR-206 and reduced expression of miRs 1, 133a, and 133b in myofiber-derived exosomes. These findings demonstrate that skeletal muscle fibers release exosomes which can exert biologically significant effects on recipient cells, and that pathological muscle conditions such as denervation induce alterations in exosomal miR profile which could influence responses to disease states through autocrine or paracrine mechanisms.

Myostatin inhibits osteoblastic differentiation by suppressing osteocyte-derived exosomal microRNA-218: A novel mechanism in muscle-bone communication

Muscle and bone are closely associated in both anatomy and function, but the mechanisms that coordinate their synergistic action remain poorly defined. Myostatin, a myokine secreted by muscles, has been shown to inhibit muscle growth, and the disruption of the myostatin gene has been reported to cause muscle hypertrophy and increase bone mass. Extracellular vesicle-exosomes that carry microRNA (miRNA), mRNA, and proteins are known to perform an important role in cell-cell communication. We hypothesized that myostatin may play a crucial role in muscle-bone interactions and may promote direct effects on osteocytes and on osteocyte-derived exosomal miRNAs, thereby indirectly influencing the function of other bone cells. We report herein that myostatin promotes expression of several bone regulators such as sclerostin (SOST), DKK1, and RANKL in cultured osteocytic (Ocy454) cells, concomitant with the suppression of miR-218 in both parent Ocy454 cells and derived exosomes. Exosomes produced by Ocy454 cells that had been pretreated with myostatin could be taken up by osteoblastic MC3T3 cells, resulting in a marked reduction of Runx2, a key regulator of osteoblastic differentiation, and in decreased osteoblastic differentiation via the down-regulation of the Wnt signaling pathway. Importantly, the inhibitory effect of myostatin-modified osteocytic exosomes on osteoblast differentiation is completely reversed by expression of exogenous miR-218, through a mechanism involving miR-218-mediated inhibition of SOST. Together, our findings indicate that myostatin directly influences osteocyte function and thereby inhibits osteoblastic differentiation, at least in part, through the suppression of osteocyte-derived exosomal miR-218, suggesting a novel mechanism in muscle-bone communication.

Human high temperature requirement serine protease A1 (HTRA1) degrades tau protein aggregates

Protective proteases are key elements of protein quality control pathways that are up-regulated, for example, under various protein folding stresses. These proteases are employed to prevent the accumulation and aggregation of misfolded proteins that can impose severe damage to cells. The high temperature requirement A (HtrA) family of serine proteases has evolved to perform important aspects of ATP-independent protein quality control. So far, however, no HtrA protease is known that degrades protein aggregates. We show here that human HTRA1 degrades aggregated and fibrillar tau, a protein that is critically involved in various neurological disorders. Neuronal cells and patient brains accumulate less tau, neurofibrillary tangles, and neuritic plaques, respectively, when HTRA1 is expressed at elevated levels. Furthermore, HTRA1 mRNA and HTRA1 activity are up-regulated in response to elevated tau concentrations. These data suggest that HTRA1 is performing regulated proteolysis during protein quality control, the implications of which are discussed.

Paired helical filaments from Alzheimer disease brain induce intracellular accumulation of Tau protein in aggresomes

Abnormal folding of tau protein leads to the generation of paired helical filaments (PHFs) and neurofibrillary tangles, a key neuropathological feature in Alzheimer disease and tauopathies. A specific anatomical pattern of pathological changes developing in the brain suggests that once tau pathology is initiated it propagates between neighboring neuronal cells, possibly spreading along the axonal network. We studied whether PHFs released from degenerating neurons could be taken up by surrounding cells and promote spreading of tau pathology. Neuronal and non-neuronal cells overexpressing green fluorescent protein-tagged tau (GFP-Tau) were treated with isolated fractions of human Alzheimer disease-derived PHFs for 24 h. We found that cells internalized PHFs through an endocytic mechanism and developed intracellular GFP-Tau aggregates with attributes of aggresomes. This was particularly evident by the perinuclear localization of aggregates and redistribution of the vimentin intermediate filament network and retrograde motor protein dynein. Furthermore, the content of Sarkosyl-insoluble tau, a measure of abnormal tau aggregation, increased 3-fold in PHF-treated cells. An exosome-related mechanism did not appear to be involved in the release of GFP-Tau from untreated cells. The evidence that cells can internalize PHFs, leading to formation of aggresome-like bodies, opens new therapeutic avenues to prevent propagation and spreading of tau pathology.

GSPE interferes with tau aggregation in vivo: implication for treating tauopathy

Tauopathies are characterized by progressive neurodegeneration caused by intracellular accumulation of hyperphosphorylated tau protein aggregates in the brain. The present study was designed to test whether a grape seed polyphenolic extract (GSPE) previously shown to inhibit tau protein aggregation in vitro could benefit tau-mediated neuropathology and behavior deficits in JNPL3 transgenic mice expressing a human tau protein containing the P301L mutation. Nine-month-old JNPL3 mice were treated with GSPE delivered through their drinking water for 6 months. We found that GSPE treatment significantly reduced the number of motor neurons immunoreactive for hyperphosphorylated and conformationally-modified tau in the ventral horns of the spinal cord identified using AT100, PHF-1, AT8, and Alz50 tau antibodies. This coincided with a drastically reduced level of hyperphosphorylated and sarcosyl-insoluble tau in spinal cord fractions. Furthermore, the reduction of tau pathology was accompanied by an improvement in the motor function assessed by a wire hang test. Collectively, our results suggest that GSPE can interfere with tau-mediated neurodegenerative mechanisms and ameliorate neurodegenerative phenotype in an animal model of tauopathy. Our studies support further evaluation of GSPE for preventing and/or treating of tauopathies in humans.

Grape derived polyphenols attenuate tau neuropathology in a mouse model of Alzheimer's disease

Aggregation of microtubule-associated protein tau into insoluble intracellular neurofibrillary tangles is a characteristic hallmark of Alzheimer's disease (AD) and other neurodegenerative diseases, including progressive supranuclear palsy, argyrophilic grain disease, corticobasal degeneration, frontotemporal dementias with Parkinsonism linked to chromosome 17, and Pick's disease. Tau is abnormally hyperphosphorylated in AD and aberrant tau phosphorylation contributes to the neuropathology of AD and other tauopathies. Anti-aggregation and anti-phosphorylation are main approaches for tau-based therapy. In this study, we report that a select grape-seed polyphenol extract (GSPE) could potently interfere with the assembly of tau peptides into neurotoxic aggregates. Moreover, oral administration of GSPE significantly attenuated the development of AD type tau neuropathology in the brain of TMHT mouse model of AD through mechanisms associated with attenuation of extracellular signal-receptor kinase 1/2 signaling in the brain.

Development of a grape seed polyphenolic extract with anti-oligomeric activity as a novel treatment in progressive supranuclear palsy and other tauopathies

A diverse group of neurodegenerative diseases - including progressive supranuclear palsy (PSP), corticobasal degeneration and Alzheimer's disease among others, collectively referred to as tauopathies - are characterized by progressive, age-dependent intracellular formations of misfolded protein aggregates that play key roles in the initiation and progression of neuropathogenesis. Recent studies from our laboratory reveal that grape seed-derived polyphenolic extracts (GSPE) potently prevent tau fibrillization into neurotoxic aggregates and therapeutically promote the dissociation of preformed tau aggregates [J. Alzheimer's Dis. (2009) vol. 16, pp. 433]. Based on our extensive bioavailability, bioactivity and functional preclinical studies, combined with the safety of GSPE in laboratory animals and in humans, we initiated a series of studies exploring the role of GSPE (Meganatural-Az(®) GSPE) as a potential novel botanical drug for the treatment of certain forms of tauopathies including PSP, a neurodegenerative disorder involving the accumulation and deposition of misfolded tau proteins in the brain characterized, in part, by abnormal intracellular tau inclusions in specific anatomical areas involving astrocytes, oligodendrocytes and neurons [J. Neuropathol. Exp. Neurol. (2002) vol. 61, pp. 33]. In this mini-review article, we discuss the biochemical characterization of GSPE in our laboratory and its potential preventative and therapeutic role in model systems of abnormal tau processing pertinent to PSP and related tauopathies.

Identification of G-protein coupled receptor kinase 2 in paired helical filaments and neurofibrillary tangles

G-protein coupled receptor kinases (GRKs) constitute a serine/threonine kinase family playing a major role in agonist-induced phosphorylation and desensitization of G-protein coupled receptors. Recently, GRK2 and GRK5 have been demonstrated to phosphorylate alpha-synuclein (Ser129) and other synuclein isoforms. We studied colocalization of GRK2, GRK5, alpha-synuclein, and tau in neurodegenerative disorders characterized by fibrillary tau inclusions and/or alpha-synuclein-enriched Lewy bodies. We found that Lewy bodies were negative for both GRK2 and GRK5 in Lewy body disease (LBD) and LBD mixed with Alzheimer disease (AD + LBD). Instead, GRK2 but not GRK5 colocalized with 40% to 50% of neurofibrillary tangles in AD + LBD and AD brains. In disorders with less prominent alpha-synucleinopathy, neuronal and glial fibrillary tau deposits known to contain distinct subsets of tau isoforms were also positive for GRK2. These deposits included tufted astrocytes and coiled bodies in progressive supranuclear palsy, astrocytic plaques in corticobasal degeneration, and Pick bodies in Pick disease. In addition, paired helical filaments isolated from AD and AD + LBD brains were found to immunogold-label for GRK2, suggesting that GRK2 could be a potential tau kinase associated with fibrillary tau. Our studies indicate that GRK2 is a novel component of neuronal and glial fibrillary tau deposits with no preference in tau isoform binding. GRK2 may play a role in hyperphosphorylation of tau in tauopathies.

Differential changes in phosphorylation of tau at PHF-1 and 12E8 epitopes during brain ischemia and reperfusion in gerbils

Cortical neurons are vulnerable to ischemic insult, which may cause cytoskeletal changes and neurodegeneration. Tau is a microtubule-associated protein expressed in neuronal and glial cells. We examined the phosphorylation status of tau protein in the gerbil brain cortex during 5 min ischemia induced by bilateral common carotid artery occlusion followed by reperfusion for 20 min to 7 days. Control brain homogenates contained 63, 65 and 68 kD polypeptides of tau immunoreactive with Alz 50, Tau 14 and Tau 46 antibodies raised against non-phosphorylated tau epitopes. Gerbil tau was also immunoreactive with some (PHF-1 and 12E8) but not all (AT8, AT100, AT180 and AT270) antibodies raised against phosphorylated tau epitopes. PHF-1 recognized a single 68 kD polypeptide and 12E8 bound the 63 kD polypeptide. During 5 min ischemia, PHF-1 immunoreactivity declined to 6%, then recovered to control levels after 20 min of blood recirculation and subsequently increased above control values 3 and 7 days later. In contrast, 12E8 immunoreactivity remained stable during ischemia and reperfusion. Our results suggest that the two phosphorylated epitopes of tau are regulated by different mechanisms and may play different roles in microtubule dynamics. They may also define various pools of neuronal/glial cells vulnerable to ischemia.

Caloric intake and Alzheimer's disease. Experimental approaches and therapeutic implications

Alzheimer's disease (AD) is a rapidly growing public health concern with potentially devastating effects. Presently, there are no known cures or effective preventive strategies. While genetic factors are relevant in early-onset cases, they appear to play less of a role in late-onset sporadic AD cases, the most common form of AD. Due to the fact that the disease typically strikes very late in life, delaying symptoms could be as good as a cure for many people. For example, it is now widely accepted that if the onset of the disease could be delayed by even 5 years, the incidence could be cut in half. Both clinical and epidemiological evidence suggests that modification of lifestyle factors such as nutrition may prove crucial to AD management given the mounting experimental evidence suggesting that brain cells are remarkably responsive to "what somebody is doing". Among other nongenetic factors influencing AD, recent studies strongly support the evidence that caloric intake may play a role in the relative risk for AD clinical dementia. Indeed, the effect of diet in AD has been an area of research that has produced promising results, at least experimentally. Most importantly, as mechanistic pathways are defined and their biochemical functions scrutinized, the evidence supporting a direct link between nutrition and AD neuropathology continues to grow. Our work, as well as that of others, has recently resulted in the development of experimental dietary regimens that might promote, attenuate or even reverse features of AD. Most remarkably, while we found that high caloric intake based on saturated fat promotes AD type Beta-amyloidosis, conversely we found that dietary restriction based on reduced carbohydrate intake is able to prevent it. This evidence is very exciting and is, in part, consistent with current epidemiological studies suggesting that obesity and diabetes are associated with a >4-fold increased risk of developing AD. The clarification of the mechanisms through which dietary restriction may beneficially influence AD neuropathology and the eventual discovery of future "mimetics" capable of anti-Beta-amyloidogenic activity will help in the development of "lifestyle therapeutic strategies" in AD and possibly other neurodegenerative disorders.

Insulin receptor deficits in schizophrenia and in cellular and animal models of insulin receptor dysfunction

Schizophrenia is associated with abnormalities in glucose metabolism that may lead to insulin resistance and a 3 fold higher incidence of type II diabetes mellitus. The goal of the present studies was to assess the role of insulin-dependent Akt signaling in schizophrenia and in animal and cellular models of insulin resistance. Our studies revealed a functional decrease in insulin receptor (IR)-mediated signal transduction in the dorsolateral prefrontal cortex (BA46) of medicated schizophrenics relative to control patients using post-mortem brain material. We found approximately 50% decreases in the content and autophosphorylation levels of IRbeta and approximately 76-78% decreases in Akt content and activity (pSer(473)-Akt). The inhibition of IRbeta signaling was accompanied by an elevated content of glycogen synthase kinase (GSK)-3 alpha and GSK-3beta without significant changes in phospho-Ser(21/9) GSK-3 alpha/beta levels. A cellular model of insulin resistance was induced by IRbeta knockdown (siRNA). As in schizophrenia, the IRbeta knockdown cells demonstrated a reduction in the Akt content and activity. Total GSK-3 alpha/beta content remained unaltered, but phospho-Ser(21/9) GSK-3 alpha/beta levels were reduced indicating a net increase in the overall enzyme activity similar to that in schizophrenia. Insulin resistance phenotype was induced in mice by treatment with antipsychotic drug, clozapine. Behavioral testing showed decreases in startle response magnitude in animals treated with clozapine for 68 days. The treatment resulted in a functional inhibition of IRbeta but the Akt activation status remained unaltered. Changes in GSK-3 alpha/beta were consistent with a net decrease in the enzyme activity, as opposed to that in schizophrenia. The results suggest that alterations in insulin-dependent Akt signaling in schizophrenia are similar to those observed in our cellular but not animal models of insulin resistance. In animal model, clozapine ameliorates IRbeta deficits at the GSK-3 alpha/beta level, which may justify its role in treatment of schizophrenia. Our studies suggest that aberrant IR function may be important in the pathophysiology of schizophrenia.

Inhibition of cyclooxygenase as potential novel therapeutic strategy in N141I presenilin-2 familial Alzheimer's disease

The present study was designed to further explore the potential cause/effect relationship between the expression of both the N141I presenilin (PS)2 mutant familial Alzheimer's disease (FAD) gene and cyclooxgenase (COX) in respect to the mechanism associated with programmed cell death in Alzheimer's disease (AD). We found that expression of mutant N141I PS2 resulting in apoptotic cell death in H4 neuronal cells coincided with >4-fold induction in the expression of the inducible form of COX-2, but not the constitutive COX-1. Moreover, we found that the expression of the N141I PS2 FAD gene strongly promoted (>2-fold) glycogen synthase kinase (GSK)-3beta activity coincidental with a reduction in the level of beta-catenin translocated from the cytoplasmic to the nuclear compartment. Most interestingly, we found that inhibition of COX-2-mediated generation of prostaglandin (PG)-E2 in H4 neuronal cells with the preferential COX-2 inhibitor nimesulide protects against N141I PS2-mediated apoptotic cell death coincidental with an inhibition of GSK-3beta activity and subsequent normalization of beta-catenin cellular distribution. The clinical relevance of this finding was confirmed by the evidence that COX-2 protein and PG-E2 concentrations were selectively increased >2-fold in the cerebral cortex of subjects harboring the N141I PS2 FAD mutation relative to wild-type PS2 AD cases. This study demonstrates for the first time that COX-2 may be a downstream effector of mutant N141I PS2-mediated apoptotic cell death and that inhibition of COX-2 may neuroprotect in AD through modulation of a GSK-3beta-beta-catenin-mediated response. The study provides support for the potential pharmacogenomic identification of N141I PS2 FAD cases that might preferentially benefit from inhibition of COX-2 during the progression of clinical dementia.

Expression of tau reduces secretion of Aβ without altering the amyloid precursor protein content in CHOsw cells

Insoluble deposits of tau and amyloid precursor protein (APP) peptides Abeta characterize Alzheimer's disease. We studied the role of tau in the metabolism of APP in cells stably expressing APP Swedish mutation (CHOsw). Transient expression of tau in CHOsw cells caused morphological changes, bundling of microtubules and perinuclear aggregation of Golgi-derived vesicles. It also reduced the secretion of Abeta(1-40) and Abeta(1-42) without altering the APP steady state levels. This was accompanied by a reduction in the gamma-secretase and an increase in the insulin degrading enzyme activities. Our results suggest that tau may play an inhibitory role in the amyloidogenic activity of APP.

Characterization of paired helical filaments by scanning transmission electron microscopy

Paired helical filaments (PHFs) are abnormal twisted filaments composed of hyperphosphorylated tau protein. They are found in Alzheimer's disease and other neurodegenerative disorders designated as tauopathies. They are a major component of intracellular inclusions known as neurofibrillary tangles (NFTs). The objective of this review is to summarize various structural studies of PHFs in which using scanning transmission electron microscopy (STEM) has been particularly informative. STEM provides shape and mass per unit length measurements important for studying ultrastructural aspects of filaments. These include quantitative comparisons between dispersed and aggregated populations of PHFs as well as comparative studies of PHFs in Alzheimer's disease and other neurodegenerative disorders. Other approaches are also discussed if relevant or complementary to studies using STEM, e.g., application of a novel staining reagent, Nanovan. Our understanding of the PHF structure and the development of PHFs into NFTs is presented from a historical perspective. Others goals are to describe the biochemical and ultrastructural complexity of authentic PHFs, to assess similarities between authentic and synthetic PHFs, and to discuss recent advances in PHF modeling.

Phosphorylation of tau at THR212 and SER214 in human neuronal and glial cultures: the role of AKT

We have reported recently that the microtubule-associated protein tau is phosphorylated in vitro by Akt, an important kinase in anti-apoptotic signaling regulated by insulin and growth factors. We also established that Akt phosphorylates tau separately at T212 and S214, two sites previously shown to be phosphorylated by glycogen synthase kinase 3beta (GSK3beta) and protein kinase A (PKA), respectively. In the present studies, we examined the relationship between Akt and T212/S214 in primary cultures of human neurons and astrocytes, and evaluated the contribution of two other kinases. In intact cells, we found a very low content of active (phospho-S473) form of Akt. We also found a low content of phospho-S214 but not phospho-T212 of tau, suggesting that only phospho-S212 may depend on Akt activity in situ. We upregulated Akt activity using two experimental models: treatment with a protein phosphatase inhibitor, okadaic acid, and transfection with a constitutively active Akt gene construct (c-Akt). Under these conditions, phosphorylation of tau at T212 and S214 was regulated independently, with little change or downregulation of phospho-T212 and dynamic upregulation of phospho-S214. Our studies revealed that Akt may influence the phospho-S214 content in a meaningful manner. They also revealed that PKA may only partially contribute to the phosphorylation of S214. In comparison, okadaic acid treatment severely depleted the content of GSK3beta and downregulated the remaining GSK3beta activity by Akt-dependent inhibition, consistent with minimal changes in phospho-T212. In summary, these results strongly suggest that in primary cultures, Akt selectively phosphorylates tau at S214 rather than T212. Our studies raise the possibility that tau S214 may participate in Akt-mediated anti-apoptotic signaling.

Phosphorylation of tau by fyn: implications for Alzheimer's disease

The abnormal phosphorylation of tau protein on serines and threonines is a hallmark characteristic of the neurofibrillary tangles of Alzheimer's disease (AD). The discovery that tau could be phosphorylated on tyrosine and evidence that Abeta signal transduction involved tyrosine phosphorylation led us to question whether tyrosine phosphorylation of tau occurred during the neurodegenerative process. In this study we determined that human tau tyr18 was phosphorylated by the src family tyrosine kinase fyn. By developing both polyclonal and monoclonal probes specific for phospho-tyr18, we found that the phosphorylation of tau at tyr18 occurred at early developmental stages in mouse but was absent in the adult. Our phosphospecific probes also revealed that paired helical filament preparations exhibited phospho-tyr18 reactivity that was sensitive to phosphotyrosine-specific protein phosphatase treatment. Moreover, immunocytochemical studies indicated that tyrosine phosphorylated tau was present in the neurofibrillary tangles in AD brain. However, the staining pattern excluded neuropil threads and dystrophic neurites indicating that tyrosine phosphorylated tau was distributed in AD brain in a manner dissimilar from other abnormally phosphorylated tau. We also found evidence suggesting that differentially phosphorylated tau existed within degenerating neurons. Our data add new support for a role for fyn in the neurodegenerative process.

A role of P301L tau mutant in anti-apoptotic gene expression, cell cycle and apoptosis

In exploring the causative role of the most common Pro(301)-to-Leu (TauP301L) tau missense mutation associated with neurodegenerative tauopathies, we examined TauP301L-mediated apoptotic cell death and the expression of a cluster of genes involved in the inhibition of apoptosis (IAPs) in human neuroblastoma SH-SY5Y cells. Our research found that the expression of TauP301L, but not wild-type tau, down regulated the expression of IAPs, including survivin, which plays a role in the mitotic spindle checkpoint. The inhibition of IAPs coincided with the activation of the pro-apoptotic caspase 3, but preceded apoptotic cell death by TUNEL. Furthermore, TauP301L altered the expression of the cell cycle regulatory proteins and induced the cell cycle arrest at G(2)/M phase. Our studies demonstrate that TauP301L downregulates the expression of genes that protect against apoptosis and regulate cell cycle progression. These results suggest a novel mechanism of apoptotic cell death in TauP301L-expressing cells that involves survivin-mediated activation of cell cycle checkpoint.

Akt/PKB kinase phosphorylates separately Thr212 and Ser214 of tau protein in vitro

Microtubule-associated protein tau contains a consensus motif for protein kinase B/Akt (Akt), which plays an essential role in anti-apoptotic signaling. The motif encompasses the AT100 double phospho-epitope (Thr212/Ser214), a specific marker for Alzheimer's disease (AD) and other neurodegenerations, raising the possibility that it could be generated by Akt. We studied Akt-dependent phosphorylation of tau protein in vitro. We found that Akt phosphorylated both Thr212 and Ser214 in the longest and shortest tau isoforms as determined using phospho site-specific antibodies against tau. Akt did not phosphorylate other tau epitopes, including Tau-1, AT8, AT180, 12E8 and PHF-1. The Akt-phosphorylated tau retained its initial electrophoretic mobility. Immunoprecipitation studies with phospho-specific Thr212 and Ser214 antibodies revealed that only one of the two sites is phosphorylated per single tau molecule, resulting in tau immunonegative for AT100. Mixed kinase studies showed that prior Ser214 phosphorylation by Akt blocked protein kinase A but not GSK3beta activity. On the other hand, GSK3beta selectively blocked Ser214 phosphorylation, which was prevented by lithium. The results suggest that Akt may be involved in AD-specific phosphorylation of tau at the AT100 epitope in conjunction with other kinases. Our data suggest that phosphorylation of tau by Akt may play specific role(s) in Akt-mediated anti-apoptotic signaling, particularly relevant to AD and other neurodegenerations.

Tau protein expression in adult bovine oligodendrocytes: functional and pathological significance

In tauopathies, overexpression of tau exon 10 is linked to degeneration and abnormal tau deposition in neurons and oligodendroglia (OLGs). To compare exon 10 expression in normal neurons and OLGs, adult bovine brain was examined for the expression of tau in gray matter and cultured OLGs isolated from white matter. Using exon-specific antibodies, we found that both types of tissues abundantly expressed exon 2 but isolated OLGs had a lower expression of exons 3 and 10 when compared to gray matter. Relative expression of exons 3 and 10 did not change significantly during the in vitro maturation of OLGs for 39 days. Using a panel of well-characterized antibodies against tau, we determined that isolated OLGs contained tau phosphorylated at the Tau-1, 12E8, and PHF-1 but not the AT8, AT100, AT180, and AT270 epitopes. Tau phosphorylation status diminished during in vitro maturation, suggesting that healthy OLG processes require regulated phosphorylation of tau at specific sites. We propose that the tau isoform profile and phosphorylation status contribute to the vulnerability of OLGs in degenerative diseases linked to overexpression of exon 10.

Morphological and biochemical correlations of abnormal tau filaments in progressive supranuclear palsy

Progressive supranuclear palsy (PSP) is characterized by specific filamentous tau inclusions present in 3 types of cells including oligodendrocytes (coiled bodies), astrocytes (tufted astrocytes), and neurons (neurofibrillary tangles; NFTs). To correlate the morphological features and biochemical composition of tau in the inclusions, we examined tau filament-enriched fractions isolated from selected brain regions. Frontal and cerebellar white matter manifested a predominance of coiled bodies. The isolated fractions contained straight, 14-nm-wide filaments of relatively smooth appearance. Caudate nucleus and motor cortex with numerous tufted astrocytes contained mostly straight, but irregular, 22-nm-wide filaments with jagged contours. Perirhinal cortex and hippocampus, rich in NFTs, contained 22-nm-wide filaments that were twisted at 80-nm intervals. Among the regions, those with tufted astrocytes showed the most heterogeneity in the ultrastructure of filaments. In all regions, isolated filaments were immunolabeled with PHF-1, Tau 46, and AT8. Fractions from all regions showed 2 PHF-1 immunoreactive bands of 64 and 68 kDa, while an additional band of 60 kDa was detected in NFT-enriched regions. All fractions, in varying extents, showed Tau-1-immunoreactive bands between 45-64 kDa. The results indicate that the 3 types of PSP tau inclusions vary in the ultrastructure although with some overlapping features. Neuronal and glial inclusions also vary in the biochemical profile of tau protein. These differences may depend on the metabolism of tau in the diseased oligodendrocytes, astrocytes, and neurons.

Structural analysis of Pick's disease-derived and in vitro-assembled tau filaments

Pick's and Alzheimer's diseases are distinct neurodegenerative disorders both characterized in part by the presence of intracellular filamentous tau protein inclusions. The tight bundles of paired helical filaments (PHFs) of tau protein found in Alzheimer's disease (AD) differ from the tau filaments of Pick's disease in their morphology, distribution, and pathological structure as identified by silver impregnation. The filaments of Pick's disease are loosely arranged in pathognomonic spherical inclusions found in ballooned neurons, whereas the tau pathology of AD is classically described as a triad of neuropil threads, neurofibrillary tangles, and dystrophic neurites surrounding and invading plaques. In this study we used the high-resolution technique of scanning transmission electron microscopy to characterize and compare the filaments found in Pick's disease with those found in AD. In addition, we determined the mass/nm length and density of arachidonic acid-induced in vitro-assembled filaments. Three morphologically distinct populations of Pick's filaments were identified but each was indistinguishable from AD-PHFs in mass/nm length and density. Filaments assembled in vitro from single isoforms were similar in mass/nm length, but less dense than AD-PHFs and Pick's disease filaments. Finally, we provide clear structural evidence that a PHF, whether found in disease or assembled in vitro, is composed of two distinct intertwined filaments.

Induction of Alzheimer-specific Tau epitope AT100 in apoptotic human fetal astrocytes

In Alzheimer's and other neurodegenerative diseases, hyperphosphorylated tau accumulates in affected neuronal and glial cells in the form of paired helical filaments (PHFs). This tau binds antibody AT100, which recognizes the double phosphorylation site (Thr212/Ser214) that is not present in normal biopsy tau. In primary cultures, highly enriched (>98%) in astrocytes of human fetal brain, three polypeptides of 52, 64, and 70 kD showed immunoreactivity with tau antibodies against non-phosphorylated epitopes, accounting for 88, 12, and <1%, respectively, of the total reactivity. All three polypeptides were phosphorylated at the PHF-1 epitope but not at the epitopes Tau-1, 12E8, AT8, and AT100. Treatment of cultures with okadaic acid resulted in apoptosis characterized by the blebbing of the plasma membrane, condensation of nuclear chromatin, and fragmentation of the nucleus. This treatment also resulted in a 3- to 5-fold increase in the content of both tau protein and phosphorylation. The increases were observed in all phosphorylation sites examined, and included the AT100 site. The AT100 site has been proposed to be generated by protein kinase B/Akt and Cdc2. Since okadaic acid can induce an AD-like hyperphosphorylated state of normal tau in primary cultures of human brain cells, a simple cellular model is available permitting study of self-aggregation of tau and phosphorylation events characteristic of neurodegeneration.

A double-labeling immunohistochemical study of tau exon 10 in Alzheimer's disease, progressive supranuclear palsy and Pick's disease

Neurofibrillary tangles (NFT), one of the histopathological hallmarks of Alzheimer's disease (AD) and progressive supranuclear palsy (PSP), and Pick bodies in Pick's disease (PiD) are composed of microtubule-associated protein tau, which is the product of alternative splicing of a gene on chromosome 17. Alternative expression of exon 10 leads to formation of three- or four-repeat tau isoforms. To study the differential expression of exon 10, we performed double-labeling immunohistochemistry of the hippocampal formation in nine AD, four PSP and three PiD cases. Cryostat sections were processed with and without formic acid (FA) treatment, and double-stained with anti-tau (Alz-50 or PHF-1) or anti-amyloid P component antibodies and one of two specific anti-exon 10 antibodies (E-10). The effect of proteinase-K treatment was also evaluated. The results suggest the following. First, in AD, E-10 immunoreactivity is present in most intracellular NFT, but not in most dystrophic neurites and neuropil threads, suggesting differential expression of tau isoforms in specific cellular domains. Second, in AD, E-10 immunoreactivity is lost or blocked in most extracellular NFT, possibly due to proteolysis. Third, in PSP, E-10 immunoreactivity is hidden or blocked in NFT and tau-positive glial inclusions, but FA treatment exposes the epitope consistent with the hypothesis that PSP inclusions contain four-repeat tau. Fourth, E-10 immunoreactivity is present in dentate fascia NFT in AD and PSP, but not in Pick bodies in the dentate fascia or other areas. The results suggest that expression of exon 10 in tau is specific for cellular domains in a disease-specific manner.

Conformation of paired helical filaments blocks dephosphorylation of epitopes shared with fetal tau except Ser199/202 and Ser202/Thr205

To determine if the high phosphate content of paired helical filaments (PHFs) in Alzheimer's disease (AD) is a result of limited access to filament phosphorylation sites, we studied in vitro dephosphorylation of intact PHFs, PHFs with filamentous structure abolished by formic acid treatment (PHF(FA)) and fetal human tau protein. Samples were treated with alkaline phosphatase for up to 24 h at 37 degrees C and then immunoblotted with eight well characterized tau antibodies, that recognize two phosphorylation-insensitive sites and six phosphorylation-sensitive epitopes at Thr181, Ser199/202, Ser202/Thr205, Thr231, Ser262/356 and Ser396/404. Intact PHFs were effectively dephosphorylated only at the two N-terminal epitopes Ser199/202 and Ser202/Thr205, with little change in electrophoretic mobility. In contrast, PHF(FA) were dephosphorylated at all epitopes, with particular effectiveness at those in the C-terminus and with significant increase in electrophoretic mobility. The fetal tau epitopes were effectively dephosphorylated except at Thr181 and Thr231 with marked increase in mobility. The extent of dephosphorylation of PHF(FA) was equal or more effective than in fetal tau, except for Thr181 that was minimally dephosphorylated in both proteins. The results indicate that intact PHFs, but not PHF(FA) or fetal tau display differential dephosphorylation of the N- and C-terminal epitopes. The results confirm that the filamentous conformation may significantly contribute to hyperphosphorylation of PHFs in the C-terminus. The filamentous conformation, however, does not limit access to two N-terminal epitopes Ser199/202 and Ser202/Thr205. The access to these sites in AD may be limited by other factors, e.g., inhibition of phosphatase binding.

Ca2+ and Mg2+ selectively induce aggregates of PHF-tau but not normal human tau

The molecular mechanism of pathological aggregation of microtubule-associated protein tau during neurodegeneration is unclear. In the present study, the in vitro effect of various metal ions on the aggregation of tau was examined using paired helical filament tau (PHF-tau) obtained from corticobasal degeneration (CBD) and Alzheimer's disease (AD) brains as well as normal human tau proteins isolated from fetal and adult brains and a recombinant system. Among the metal ions tested, Ca2+ and Mg2+ effectively induced formation of approximately 340 kD aggregates of PHF-tau but not normal tau proteins as determined by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis and immunoblotting. Al3+ and Fe2+ precipitated both PHF-tau and normal tau protein as SDS-insoluble pellets. The other metal ions examined (Cu2+, Zn2+, and Li+) were inactive and caused neither aggregation nor precipitation of any tau protein. Intermixing experiments using PHF-tau and various normal tau preparations showed that the 340-kD aggregates induced by Ca2+ contained PHF-tau but not normal tau regardless whether unmodified (recombinant) or highly phosphorylated (fetal brain) tau proteins were used. The present results suggest that post-translational modifications other than the fetal-type phosphorylation are required for Ca2+- and Mg2+-dependent aggregation of PHF-tau and that the regional elevation of these ions may trigger pathological deposition of PHF-tau in certain neurodegenerative disorders.

Assembled tau filaments differ from native paired helical filaments as determined by scanning transmission electron microscopy (STEM)

Paired helical filaments (PHF) are abnormal, approximately 20-25-nm wide periodically twisted filaments, which accumulate in Alzheimer's disease (AD) brain and other neurodegenerative disorders, including corticobasal degeneration (CBD). PHF are primarily composed of highly phosphorylated tau protein. However, both phosphorylated and non-phosphorylated forms of tau are able to assemble in vitro into filaments similar in the ultrastructural appearance to PHF. In the present study, filaments were assembled in vitro from unmodified recombinant human tau and the physical mass per unit length of filaments and the mass density were determined using scanning transmission electron microscopy (STEM). Two general types of filaments were observed. One type was composed of 11.4 nm-wide, 10-75 nm long, frequently twisted and PHF-like filaments, with a mass per unit length (44 kDa/nm) approximately one third of that observed in isolated AD filaments. The other were straight filaments, approximately 6.8-nm wide and 0.2-2 microm long, which often formed parallel clusters of two or more filaments. Triple clusters were 19. 2-nm wide and had a mass per unit length (70 kDa/nm) approximately two thirds of that seen in isolated AD filaments. Despite different morphology, both twisted and straight filaments had mass densities between 0.48-0.55 kDa/nm3. These values are significantly higher than those reported for PHF found either in AD (0.40 kDa/nm3) or CBD (0.33 kDa/nm3). These results suggest that the packing of tau differs in vivo from that observed in vitro and that specific tau isoform content, elongation of tau molecules by phosphorylation or other factors may be required to reproduce pathological assembly. Therefore mass density determinations appear to be an important criterion in comparing various filaments.

Ubiquitin immunoreactivity of paired helical filaments differs in Alzheimer's disease and corticobasal degeneration

To establish whether there is a relationship between ubiquitination and ultrastructural appearance of filaments, we compared the ubiquitin immunoreactivity of paired helical filaments (PHFs) in Alzheimer's disease (AD) and corticobasal degeneration (CBD). PHFs in these disorders share a limited similarity since filaments in CBD are wider and twisted at longer intervals than those in AD, and also display less ultrastructural stability. Preparations enriched in SDS-soluble filaments were isolated from AD and CBD brains and subjected to tau and ubiquitin immunogold labeling. Both preparations contained mostly dispersed individual PHFs, which labeled for the amino and carboxyl termini of tau. Immunolabeling of ubiquitin was variable, however, being more intense in AD than CBD samples. SDS-insoluble filaments were prepared from PHFs by boiling in the presence of SDS and 2-mercaptoethanol and collected by sedimentation. In both disorders, the pellets contained highly aggregated and bundled filaments, which were devoid of the amino but not the carboxyl terminal region of tau. Again, ubiquitin labeling was more intense in AD than CBD filaments. The present results suggest that ubiquitination has limited influence on SDS solubility, aggregation and bundling of PHFs; however, it may be one of the factors responsible for the ultrastructural variability and/or stability of filaments.

Paired helical filaments in corticobasal degeneration: the fine fibrillary structure with NanoVan

Paired helical filaments (PHF) composed of hyperphosphorylated tau proteins are characteristic findings in neurodegenerative disorders, including Alzheimer's disease (AD) and corticobasal degeneration (CBD). The filaments in CBD differ from those in AD by a reduced number of tau isoforms and less stable ultrastructure. To further compare the ultrastructure of both filaments, we employed a novel staining reagent, NanoVan, as well as aurothioglucose and uranyl acetate. With commonly used uranyl acetate, both kinds of filaments appeared as twisted ribbons 15-20-nm and 21-23-nm wide, respectively, without significant internal substructure. With application of aurothioglucose, only few structural details were apparent. With NanoVan, AD filaments showed similar structure to that with uranyl acetate but CBD filaments displayed a highly heterogeneous appearance consistent with the dissociation of the 20-25-nm-wide filaments along two longitudinal axes. This was evident by the presence of thinner, 12-13-nm-wide filaments and filaments that splayed into two 20-25-nm-wide components at one or both ends. Moreover, detection of a prominent, 7-8-nm-wide axial region distinguished up to four protofilaments per one filament. Each protofilament appeared to contain two 3-5-nm-wide fibrils separated by an approximately 1-nm-wide axial region. The results suggest that 3-5-nm fibrils are the smallest structural subunits of filaments in CBD and that NanoVan may be an unique reagent in detecting eight-fibril organization in these less stable filaments.

Tau released from paired helical filaments with formic acid or guanidine is susceptible to calpain-mediated proteolysis

Paired helical filaments (PHFs), a characteristic neuropathologic finding in Alzheimer's disease brain, are abnormal fibrillary forms of hyperphosphorylated tau (PHF-tau), which have been shown to be highly resistant to calpain digestion. Either excessive phosphorylation or fibrillary arrangement of tau proteins in PHFs may play a role in proteolytic resistance by limiting access to calpain recognition/digestion sites. To determine the contribution of the fibrillary conformation, isolated PHFs were subjected to treatment with either formic acid or guanidine. Both procedures effectively abolished the fibrillary structure of PHF but preserved PHF-tau immunoreactivity using a panel of antibodies that recognize nonphosphorylated and phosphorylated epitopes. These treatments also significantly increased the sensitivity of PHF-tau polypeptides to calpain proteolysis as shown by significant decreases in the half-life (t(1/2)) from the infinite with native PHF to 44 min and 4.4 min in formic acid- or guanidine-treated samples, respectively. In contrast, the sensitivity of normal fetal tau (3.4 min) was either decreased (5.9 min) or unaffected (3.6 min) by similar treatment. Our results indicate that after guanidine treatment, the sensitivity of PHF to calpain resembles that of fetal tau. These results strongly suggest that the fibrillary structure of PHF-tau, rather than hyperphosphorylation, is the major factor responsible for the resistance of abnormal filaments to calpain-mediated proteolysis.

Ultrastructural instability of paired helical filaments from corticobasal degeneration as examined by scanning transmission electron microscopy

Paired helical filaments (PHFs) accumulate in the brains of subjects affected with Alzheimer's disease (AD) and certain other neurodegenerative disorders, including corticobasal degeneration (CBD). Electron microscope studies have shown that PHFs from CBD differ from those of AD by being wider and having a longer periodicity of the helical twist. Moreover, PHFs from CBD have been shown to be primarily composed of two rather than three highly phosphorylated polypeptides of tau (PHF-tau), with these polypeptides expressing no exons 3 and 10. To further explore the relationship between the heterogeneity of PHF-tau and the appearance of abnormal filaments, the ultrastructure and physical parameters such as mass per unit length and dimensions were compared in filaments from CBD and AD using high resolution scanning transmission electron microscopy (STEM). Filament-enriched fractions were isolated as Sarcosyl-insoluble pellets and for STEM studies, samples were freeze-dried without prior fixation or staining. Ultrastructurally, PHFs from CBD were shown to be a heterogeneous population as double- and single-stranded filaments could be identified based on their width and physical mass per unit length expressed in kilodaltons (kd) per nanometer (nm). Less abundant, double-stranded filaments had a maximal width of 29 nm and a mass per unit length of 133 kd/nm, whereas three times more abundant single-stranded filaments were 15 nm wide and bad a mass per unit length of 62 kd/nm. Double-stranded filaments also displayed a distinct axial region of less dense mass, which appeared to divide the PHFs into two protofilament-like strands. Furthermore, these filaments were frequently observed to physically separate along the long axis into two single strands or to break longitudinally. In contrast, PHFs from AD were ultrastructurally stable and uniform both in their width (22 nm) and physical mass per unit length (104 kd/nm). The ultrastructural features indicate that filaments of CBD and AD differ both in stability and packing of tau and that CBD filaments, composed of two distinct protofilaments, are more labile under STEM conditions. As fixed and stained filaments from CBD have been shown to be stable and uniform in size by conventional transmission electron microscopy, STEM studies may be particularly suitable for detecting instability of unstained and unfixed filaments. The results also suggest that molecular heterogeneity and/or post-translational modifications of tau may strongly influence the morphology and stability of abnormal filaments.

Binding of Alz 50 depends on Phe8 in tau synthetic peptides and varies between native and denatured tau proteins

Alz 50 is a monoclonal antibody that in Western blotting analysis recognizes both normal tau as well as hyperphosphorylated tau proteins associated with paired helical filaments (PHF-tau) in Alzheimer disease (AD). Within tissue sections of AD brain, however, Alz 50 immunolabels only PHF, which suggests that the antibody recognizes a conformational epitope. Using competitive enzyme-linked immunosorbent assay, we demonstrate that Alz 50 binds to tau synthetic peptides with low affinity (KD between 0.27 to 2.7 x 10(-5) M) and that the binding is specific for the RQEF sequence corresponding to N-terminal residues 5-8 of tau. The Alz 50 epitope appears to be largely dependent on Phe8, a strongly hydrophobic amino acid residue, since the substitution of Phe8 with Ala8 in the synthetic peptide abolishes Alz 50 binding. The effects of tau conformation on Alz 50 binding were studied with various normal tau proteins with either low or high phosphate content (adult vs. fetal) and PHF-tau proteins. The normal tau fractions were isolated from both adult and fetal human brains using affinity chromatography (native form) and heat/perchloric acid treatments (denatured form). PHF-tau was isolated as Sarcosyl-insoluble fraction. With competitive ELISA, the denatured form of normal tau (fetal and adult) bound Alz 50 with the same high affinity as did PHF-tau (KD between 1.3 to 1.8 x 10(-7) M). In contrast, the native form of tau from either brain was unable to fully compete for Alz 50 and at most only 50% of the Alz 50 binding sites in native tau were occupied. These results suggest that native tau may exist either in complexes with other proteins or in a form of dimers/oligomers, in which only some N-termini are available for binding (e.g. head-to-tail assembly). The results also suggest that denaturation rather than phosphorylation of tau has more significant effect on interactions of tau with Alz 50.

Calpain-induced proteolysis of normal human tau and tau associated with paired helical filaments

The major components of neurofibrillary tangles (NFT) in Alzheimer's disease are bundles of paired helical filaments (PHF) which are primarily composed of highly phosphorylated tau proteins (PHF-tau). To further understand the mechanism of PHF accumulation in NFT, we examined the calpain-induced proteolysis of highly purified and primarily non-aggregated PHF and normal tau proteins with various contents of phosphate isolated from either fetal (F-tau) or adult human brain (N-tau). The extent of proteolysis was determined by decreases in tau immunoreactivity using Western-blot analysis and a panel of site-specific tau antibodies (Alz 50, Tau-2, Tau 14, Tau-1, AT8, E-11, AH-1 and PHF-1). We found that full-size polypeptides of N-tau and F-tau were similarly and rapidly proteolyzed in vitro by calpain (calpain II, 3.3 units/mg protein) during a 10-min incubation at 30 degrees C, and that their half lives (t1/2) were 1.5 min and 1.8 min, respectively. Analysis of immunoblots suggests that full-length polypeptides of tau are first degraded into large fragments similar in size to that generated endogenously, then into smaller fragments. Since both endogenous and in-vitro-generated tau fragments retained N-terminal epitopes, the results suggest that most of the calpain-sensitive sites may be located in the C-terminal half of the tau molecule. In contrast, PHF were extremely resistant to degradation and only a fivefold higher concentration of calpain (16.7 units/mg protein) induced partial proteolysis of PHF. A major calpain-generated fragment was a 45-kDa polypeptide derived from the C-terminal region of PHF-tau, which forms a core of filaments. The results suggest that the inaccessibility of potential calpain-digestion sites in the filament core could contribute to the resistance of PHF to calpain and subsequently lead to the accumulation of PHF in Alzheimer's disease. The results also suggest that hyperphosphorylation of tau may be marginally involved in the resistance of PHF to degradation by calpain. Ultrastructural examination revealed that, in contrast to previous studies with trypsin, calpain did not alter the morphologic appearance of filaments; after incubation with calpain, the majority of PHF remained short and disperse and the number of PHF aggregated into NFT-like clusters was not significantly increased. The results suggest that the role of calpain in promoting the aggregation and clustering of filaments is limited.

Differential expression of exons 10 and 11 in normal tau and tau associated with paired helical filaments

Antibodies were raised to two synthetic peptides with amino acid sequences encoded by a variable region of exons 10 and 11 of the tau gene. The affinity-purified antibodies, designated E-10 and E-11, were used to determine whether PHF-tau and normal tau differ in variants containing three or four repeats in the microtubule-binding domain, respectively. Normal adult human brain was shown by gel electrophoresis to contain six isoforms of tau. All of the isoforms reacted with E-11, whereas only four of them with slower electrophoretic mobility were recognized by E-10. Fetal brain tau was readily recognized by E-11 but reacted poorly with E-10. In PHF preparations, E-11 bound to all three polypeptides of PHF-tau of 68 kD, 64 kD, and 60 kD and reacted intensely with a material smearing from the top of the gel to about the 50-kD region. In contrast, E-10 only weakly recognized the two higher molecular weight PHF-tau polypeptides of 68 kD and 64 kD, as well as smeared material, and the binding was not affected by phosphatase treatment. Using recombinant tau with four repeats as a reference, the immunoreactivity of E-10 with PHF-tau was estimated to be approximately 5% of that of E-11. By comparison, the immunoreactivity of E-10 with four isoforms of normal tau was comparable to that of E-11. These results indicate that the ratio of three vs. four repeat variants in PHF-tau is higher than in normal tau and suggest that Alzheimer disease may be associated with the disproportional expression of fetal (or juvenile) forms of tau. Alternatively, the weak reactivity of PHF-tau with E-10 antibody could be due to post-translational modifications other than phosphorylation.

Epitope expression and hyperphosphorylation of tau protein in corticobasal degeneration: differentiation from progressive supranuclear palsy

Corticobasal degeneration (CBD) is a rare, progressive neurological disorder characterized by widespread neuronal and glial accumulation of abnormal tau protein. Using immunohistochemistry we analyzed tau epitope expression and phosphorylation state in CBD and compared them to cytoskeletal changes in Alzheimer's disease (AD) and progressive supranuclear palsy (PSP). Epitopes spanning the entire length of the tau protein were present in CBD inclusions. An antibody against the alternatively spliced exon 3 did not recognize cytoskeletal lesions in CBD, but did in AD and PSP. Tau epitopes from each region of the molecule were present in cytoskeletal inclusions in CBD, including gray matter astrocytic plaques, gray and white matter threads, and oligodendroglial inclusions. As in AD, tau from CBD was highly phosphorylated. Antibodies that recognized phosphorylated tau epitopes reacted with material from CBD in a highly phosphatase-dependent manner. Again, all types of inclusions contained phosphorylated epitopes. We conclude that abnormal tau protein in CBD comprises the entire tau molecule and is highly phosphorylated, but is distinguished from AD and PSP by the paucity of epitopes contained in the alternatively spliced exon 3.

Ultrastructure and biochemical composition of paired helical filaments in corticobasal degeneration

Corticobasal degeneration (CBD) is a neurodegenerative disorder associated with extensive cytoskeletal abnormalities. These include tau-positive neuropil threads and grains, ballooned or swollen neurons, neurofibrillary tangles, and glial inclusions. Given the presence of tau-positive structures in CBD, we investigated whether abnormalities in tau proteins associated with CBD were similar to those in Alzheimer's disease (AD). Fractions of abnormal tau proteins were isolated as Sarkosyl-insoluble pellets. By electron microscopic examination, the fraction from CBD contained twisted filaments that differed from paired helical filaments of AD. In CBD, filaments were shorter in length, rarely longer than 400 nm, 10 to 20% wider in the maximum and minimum widths (26 to 28 nm and 13 to 14 nm, respectively), and the periodic twist (169 to 202 nm) was twice as long as that in AD. Immunogold labeling with a panel of tau-reactive antibodies (Alz 50, Tau 14, AH-1, E-11, PHF-1, and Tau 46) showed no apparent differences in the pattern of tau immunoreactivity between filaments of CBD and AD. Western blots revealed that polypeptides of abnormal tau were present in both fractions; however, only two polypeptides (68 and 64 kd) were present in CBD as compared with three (68, 64, and 60 kd) in AD. Both of these polypeptides were reactive with additional antibodies (E-9, Tau-1 after dephosphorylation, AT8, and NP8). Only one polypeptide (68 kd) bound an antibody to adult-specific tau sequence encoded by exon 2, but neither was reactive with antibodies to adult-specific sequences encoded by exons 3 and 10. The results suggest that abnormalities in the number and heterogeneity of isoforms of tau may be one of the factors contributing to ultrastructural differences in pathological filaments of CBD and AD.

Tau immunoreactivity and SDS solubility of two populations of paired helical filaments that differ in morphology

To further understand the processes that lead to the formation of neurofibrillary tangles from paired helical filaments (PHF) in Alzheimer brains, we studied two morphologically distinct fractions of PHF separated on sucrose density gradient. In a fraction with mostly short and non-aggregated PHF, the majority of filaments could be solubilized in SDS. In a fraction containing primarily PHF aggregated into clusters or bundles, sometimes resembling neurofibrillary tangles, filaments were less soluble in SDS. Immunogold labelling with a panel of tau-immunoreactive antibodies demonstrated that N-terminal epitopes of tau were preserved in the short filaments, but were reduced or absent in aggregated filaments. In contrast, C-terminal epitopes were present in both fractions. Furthermore, the accessibility of the microtubule-binding domain to immunolabelling was markedly impaired in short and non-aggregated filaments compared to aggregated filaments. These results are consistent with proteolytic degradation of the N-terminal epitopes and preservation of the C-terminal epitopes and the microtubule-binding domain of tau in the aggregated filaments. Partial proteolysis may be involved in the generation of aggregated PHF in neurofibrillary tangles.

Mass and physical dimensions of two distinct populations of paired helical filaments

We studied the ultrastructure of two fractions of paired helical filaments (PHF) from Alzheimer brains separated on sucrose density gradient. Fraction A2 (1M sucrose) contained filaments which were short in length and did not aggregate while those in fraction AL2 (1/1.5 M sucrose interface) were mostly aggregated. By scanning transmission electron microscopy, PHF in fraction A2 had significantly more mass per nm length of filament (107-120 kD/nm) than those in fraction AL2 (79-85 kD/nm), and they were also wider in their maximum and minimum widths but did not differ in their periodicity. Differences in mass and dimensions between two morphologically distinct populations of PHF suggest that a partial proteolysis may be involved in the generation of the aggregated population of PHF. The results suggest that a similar process may be active in the formation of neurofibrillary tangles.

Phosphate analysis and dephosphorylation of modified tau associated with paired helical filaments

We performed phosphate analysis of tau proteins isolated from normal human brain, tau proteins associated with paired helical filaments (PHF-tau), and Alzheimer tau not associated with PHF. These tau fractions were of high purity. Normal and Alzheimer tau were purified by heat treatment, acid extraction and calmodulin-affinity chromatography with or without HPLC. Fractions containing primarily PHF-tau polypeptides of 60, 64 and 68 kDa and their degraded fragments were purified either on a sucrose density gradient as filaments (PHF) or by heat treatment and acid extraction as amorphous proteins (PHF-tau). PHF and PHF-tau were found to contain 6-8 mol phosphate/mol protein while normal and Alzheimer tau proteins contained 1.9 and 2.6 mol phosphate/mol protein, respectively. Upon 2-h incubation with alkaline phosphatase, PHF lost two of the phosphate groups without apparent changes in the stability and morphology of PHF. The released phosphate originated from the N-terminal half of PHF-tau as determined by immunoblotting with antibodies to epitopes blocked by phosphorylation. Tau-1 and E-2, and by a prominent shift in the electrophoretic mobility of some fragments of PHF-tau. The shift in mobility was not observed with the C-terminal fragments of 25-26 kDa, which retained the epitope to Tau 46. The results suggest that the phosphorylation sites not affected by phosphatase may be located in the 25-26 kDa C-terminal region of PHF-tau and may play a role in structural stability of PHF.

Immunocytochemistry of neurofibrillary tangles with antibodies to subregions of tau protein: identification of hidden and cleaved tau epitopes and a new phosphorylation site

Antibodies to multiple epitopes spanning the length of the tau molecule were used to study Alzheimer neurofibrillary tangles (NFT) using immunocytochemical methods and several different methods of fixation and tissue processing, including staining of vibratome sections, hydrated autoclaving of paraffin sections and immunofluorescence of NFT isolated from fresh brain tissue. Smears and sections were pretreated with trypsin and/or phosphatase to further characterize antibody binding. In tissue fixed briefly in periodate-lysine-paraformaldehyde, tau immunoreactivity was detected in astrocytes, but only a few tau epitopes were detected in NFT with this fixation method. In contrast, all tau epitopes were detected in NFT in tissue fixed in formaldehyde for prolonged periods of time. In the hippocampus, the number of NFT detected in the dentate fascia was in proportion to the duration of dementia, as we previously noted. Dentate fascia NFT were intracellular (i-NFT) and were reactive with antibodies recognizing epitopes in both the carboxy- and amino-terminal regions of tau, but not the microtubule-binding domain of tau, suggesting that microtubule-binding domain epitopes are hidden in i-NFT. In contrast, NFT in the subiculum and layer II of the parahippocampal cortex were mostly extracellular (e-NFT), especially in severe cases of long duration. e-NFT were immunoreactive with antibodies to the microtubule-binding domain, but only weakly reactive with antibodies to carboxy- or amino-terminal epitopes, suggesting that e-NFT may contain fragments of tau. In both isolated NFT and NFT in sections, amino-terminal epitopes, including the Alz-50 epitope, were sensitive to trypsin proteolysis, which suggests that the lack of staining of e-NFT by antibodies to the amino-terminal regions of tau is due to proteolysis. Antibodies reactive with amino-terminal epitopes also stained fewer NFT following hydrated autoclaving, while those reacting with the carboxy half of tau stained more NFT after hydrated autoclaving. Thus, although carboxy-terminal regions are not detected in e-NFT, they are probably masked, rather than proteolytically cleaved, since they can be revealed by hydrated autoclaving. Finally, phosphatase treatment of isolated NFT revealed enhanced immunostaining not only with Tau-1, as in previous studies demonstrating abnormal phosphorylation of tau proteins in NFT, but also with an antibody to exon 2, which reveals yet another phosphorylation site in tau of NFT.

The N terminal region of human tau is present in Alzheimer's disease protein A68 and is incorporated into paired helical filaments

Antibody (Ab) E-1 was raised to the amino terminus (19 to 33 amino acid residues) of human tau. It recognized Alzheimer's disease proteins A68 (MW 60, 64, 68 kd), labeled paired helical filaments, and had no reactivity with tau from rat, mouse, and bovine brains. The results indicate that the N terminus of tau is incorporated in A68 proteins and paired helical filaments and that human tau proteins contain species-specific amino acid sequences.

Abnormal tau proteins from Alzheimer's disease brains. Purification and amino acid analysis

Abnormal tau proteins (PHF-tau) were isolated from Alzheimer's disease brains by treatment of paired helical filament enriched-fractions with perchloric acid and boiling of the acid precipitable fraction with beta-mercaptoethanol. These proteins were purified further by a second perchloric acid treatment. The purified PHF-tau proteins were soluble in buffers devoid of sodium dodecyl sulfate. However, they were similar to the abnormal tau extracted from paired helical filaments with sodium dodecyl sulfate, also named A68, in molecular mass (68, 64, and 60 kDa), isoelectric point (pI 5.5-6.5), reactivity with anti-tau antibodies, and in requirement for alkaline phosphatase treatment to bind the Tau-1 antibody. Compared to normal tau, the soluble PHF-tau contained 100% more glycine and 35% less lysine residue. The results suggest that besides phosphorylation other types of modification may be involved in differentiating PHF-tau from normal tau.

Ubiquitin-immunoreactive dystrophic neurites in Down's syndrome brains

Ubiquitin-immunoreactivity was studied in Down's syndrome brains ranging in age from two days to sixty years. Numerous randomly distributed ubiquitin-immunoreactive dot-like structures in the white matter were shown to correspond to granular degeneration of myelin. Granular degeneration of myelin was first detected at age 21 and increased thereafter with age. Other larger and more coarsely granular ubiquitin-immunoreactive structures, most numerous in the middle and upper cortical layers, were consistent with dystrophic neurites. Immunoelectron microscopy demonstrated that the dystrophic neurites contained non-filamentous, membranous, dense bodies. In Down's syndrome, ubiquitin-immunoreactive dystrophic neurites were first detected at age six in the hippocampus, and were consistently more numerous in comparison to age-matched control subjects. In the presence of amyloid, either as diffuse or as compact deposits, ubiquitin-immunoreactive dystrophic neurites frequently formed aggregates consistent with senile plaques. Although apparently independent events, these data suggest that amyloid deposition is associated with local accentuation of ubiquitin-immunoreactive neuritic dystrophy. In addition, since dystrophic neurites appeared substantially earlier in the grey matter in Down's syndrome than in age-matched normals, this may be further evidence that selective aspects of aging are accelerated in Down's syndrome.

Structural stability of paired helical filaments requires microtubule-binding domains of tau: a model for self-association

Highly purified and SDS-soluble paired helical filaments (PHFs) were immunogold labeled and immunoblotted with antibodies to tau: Tau 14 (N-terminal half), AH-1 (microtubule-binding domain), and Tau 46 (C-terminal end). The main component of PHFs was modified tau of 68, 64, and 60 kd, also called A68 or PHF-tau. Trypsin digestion reduced the maximum width of PHFs by 10%-20%, increased aggregation of filaments, and abolished the binding of Tau 14, but had no effect on the binding of AH-1. The smallest tau-reactive tryptic fragments were 13 and 7-8 kd, positive with AH-1, and negative with Tau 46. Our results and the model of Crowther and Wischik suggest that by self-association and anti-parallel arrangement of the microtubule-binding domains, PHF-tau forms the backbone of PHFs.

Ubiquitin immunoreactivity in kuru plaques in Creutzfeldt-Jakob disease

Cerebellar kuru plaques in 2 cases of Creutzfeldt-Jakob disease were studied immunohistochemically. Similar to cerebellar senile plaques in Alzheimer's disease, many kuru plaques contained ubiquitin-positive, tau-negative small granular elements, presumably representing dystrophic neurites. Our results suggest that similar mechanisms are involved in neuritic changes in cerebellar plaques in Creutzfeldt-Jakob and Alzheimer's diseases despite differences of amyloid proteins in the plaques.

Ubiquitin immunoreactive structures in normal human brains. Distribution and developmental aspects

Ubiquitin-immunoreactive structures in normal human brains ranging in age from 2 months to 91 years were studied with light and electron microscopy. Antibodies to ubiquitin immunostained structures in both neurons and glia. In the cerebrum, ubiquitin-immunoreactive, coarsely granular structures were most consistent with dystrophic neurites. They were most numerous in middle and upper cortical layers, especially lamina II of the entorhinal cortex and the cortical and accessory basal nuclei of the amygdala. Dystrophic neurites were first detected in brains of young adults, increased with age, and were numerous in the oldest brains. One of the normal elderly subjects had a small number of senile plaques with dystrophic neurites similar to those in the gray matter of the other brains, except for their location adjacent to amyloid deposits. With immunoelectron microscopy, dystrophic neurites were nonmyelinated neuronal processes containing dense, lamellar bodies, and finely granular material. White matter consistently had more immunoreactive structures than gray matter at all ages. The immunoreactive structures in white matter were smaller, less coarsely granular "dot-like" structures. With immunoelectron microscopy, dot-like structures were composed of dense inclusions within glial cells and focal swellings in myelin lamellae containing heterogeneous dense material. Only rarely were axons immunostained. Axonal spheroids in the basal ganglia, substantia nigra, and dorsal medulla were ubiquitin-immunoreactive. Spheroids were detected in these locations as early as the second decade, and they increased in number with age. A few dystrophic axons could be detected in spinal nerve roots of the oldest subjects. Other ubiquitin-immunoreactive structures included nuclei of small granular neurons, especially those in lamina II of the neocortex of the youngest brains; round cytoplasmic inclusions in tanycytes of all brains; and intranuclear Marinesco bodies in the substantia nigra and eosinophilic cytoplasmic inclusions in inferior olivary neurons in the oldest brains. These results demonstrate the spectrum of ubiquitinated structures in normal brains and suggest that progressive axonal dystrophy may be a more common age-related pathologic alteration of the brain than formerly recognized.

Mapping of the Alz 50 epitope in microtubule-associated proteins tau

Alz 50 and seven other monoclonal antibodies have been shown to react with both tau and Alzheimer brain proteins of molecular mass 60-70 kDa. The location of some of the epitopes of these antibodies (Alz 50, Tau-2, NP14, Ab 636.7) on the tau molecule is unknown, whereas those of others (Tau 60, Tau 14, Tau-1, Tau 46) have recently been demonstrated in fetal human tau at amino acid residues 60-72, 83-120, 131-140, and 315-352. To determine the location of the unknown epitopes, human tau was digested with chymotrypsin and trypsin, and the bovine microtubule fraction was incubated with chymotrypsin. Comparison of the immunoblots of chymotryptic digested tau with those of untreated preparations showed that the Alz 50 epitope was more sensitive than other tau epitopes to proteolysis. Cleavage of a 3-4 kDa polypeptide from the periphery of tau was sufficient to remove the Alz 50 epitope, but not the epitopes of Tau 46 (C-end) or Tau 60 (N-end). The distribution of the Alz 50 epitope in endogenously degraded, chymotrypsin or trypsin digested tau fragments was different from that of the Tau 46 epitope known to be located within 38 residues from the C-terminus of the tau molecule. Based on these observations Alz 50 epitope was considered to be located within 3-4 kDa of the N-terminus of tau. A comparison of immunoblots of different tau-reactive antibodies showed similarities between Tau 60 and Tau-2, and between Tau 14, Tau-1, NP14, and Ab 636.7.(ABSTRACT TRUNCATED AT 250 WORDS)

Alzheimer disease proteins (A68) share epitopes with tau but show distinct biochemical properties

Alz 50, a monoclonal antibody raised against Alzheimer brain homogenate, reacts with neurofibrillary tangles, microtubule-associated proteins tau, and Alzheimer brain proteins of molecular weight 70-60 kDa (A68). To study the relationship between A68 and normal human tau we compared the biochemical properties of these proteins and tested the reactivity of A68 with eight antibodies (Alz 50, Tau 60, Tau-2, Tau 14, Tau-1, Ab 636.7, NP14, Tau 46) that bind to various regions of tau molecule. On Western blots, all tau-reactive antibodies, except Tau-1, recognized A68. Pretreatment with alkaline phosphatase was required for the Tau-1 binding to A68. A68 consisted of three polypeptides of 68, 64, and 60 kDa, while tau contained 4-6 polypeptides of 50-65 kDa. A68 was less heterogenous than tau in the number of pI variants on two-dimensional gels. All A68 variants were more acidic (pI 5.5-6.5) than human tau (pI 6.5-8.5). Phosphatase treatment had only a minor effect on the pI and mobility of A68. Limited proteolysis of A68 with trypsin or chymotrypsin generated large fragments of 56-66 kDa (chymotrypsin) and 40-45 kDa (trypsin). While none of the fragments was recognized by Alz 50, the chymotryptic fragments were reactive with all the other tau antibodies, and the tryptic fragments were positive with five of the antibodies (Tau 14, Tau-1, Ab 636.7, NP14, and Tau 46). The peptide maps of A68 differed from that of tau in the number and the size of the peptide fragments. The differences in biochemical properties of these proteins and the sharing multiple epitopes suggest that A68 is a modified form of tau. The modification in part may be due to phosphorylation, although other changes rendering different isoelectrical properties and susceptibility to proteases need to be considered. The removal of the Alz 50 epitope by a cleavage of a 2-3 kDa fragment which does not contain the most C-terminal epitope (Tau 46) indicates that the Alz 50 epitope is located at the N-terminal periphery of the A68 molecule.