Casein kinase 1 delta is associated with pathological accumulation of tau in several neurodegenerative diseases

Neurobiology of Aging 21 (2000) 503–510

www.elsevier.com/locate/neuaging

Casein kinase 1 delta is associated with pathological accumulation of tau in several neurodegenerative diseases夞 Claudia Schwaba,*, Anthony J. DeMaggiob, Nupur Ghoshalc, Lester I. Binderc, Jeff Kuretd, Patrick L. McGeera a

Kinsmen Laboratory of Neurological Research, University of British Columbia, Vancouver, B.C., Canada b ICOS Corporation, Bothell, WA, USA c Department of Cell and Molecular Biology, Cognitive Neurology, and Alzheimer’s Disease Center, Northwestern Medical School, Chicago, IL, USA d Department of Medical Biochemistry, Ohio State University, Columbus, OH, USA Received 20 September 1999; received in revised form 24 January 2000; accepted 10 February 2000

Abstract The distribution of casein kinase 1 ␦ (Cki␦) was studied by immunohistochemistry and correlated with other pathological hallmarks in Alzheimer’s disease (AD), Down syndrome (DS), progressive supranuclear palsy (PSP), parkinsonism dementia complex of Guam (PDC), Pick’s disease (PiD), pallido-ponto-nigral degeneration (PPND), Parkinson’s disease (PD), dementia with Lewy bodies (DLB), amyotrophic lateral sclerosis (ALS), and elderly controls. Cki␦ was found to be associated generally with granulovacuolar bodies and tau-containing neurofibrillary tangles in AD, DS, PSP, PDC, PPND, and controls, and Pick bodies and ballooned neurons in PiD. It was not associated with tau-containing inclusions in astroglia and oligodendroglia in PPND, PSP, and PDC. It was also not associated with tau-negative Lewy bodies in PD and DLB, Hirano bodies in PDC, Marinesco bodies in PD, AD, and controls and “skein”-like inclusions in anterior motor neurons in ALS. The colocalization of the kinase Cki␦ and its apparent substrate tau suggests a function for Cki␦ in the abnormal processing of tau. © 2000 Elsevier Science Inc. All rights reserved. Keywords: Casein kinase 1 delta; Alzheimer’s disease; Pick’s disease; Parkinson’s disease; Parkinsonism dementia complex of Guam; Pallido-ponto-nigral degeneration; Granulovacuolar bodies; Neurofibrillary and glial tangles; Lewy bodies; Immunocytochemistry

1. Introduction Aberrantly phosphorylated tau accumulates in such disorders as Alzheimer disease (AD), Down syndrome (DS), progressive supranuclear palsy (PSP), corticobasal degeneration, and parkinsonism dementia complex of Guam (PDC). The hyperphosphorylated tau forms both straight and paired helical filaments that accumulate within neurofibrillary tangles (NFT), neuropil threads, and dystrophic neurites (DN) in amyloid-containing plaques. In Pick’s disease (PiD), abnormal tau accumulates in Pick bodies and ballooned neurons. In several tauopathies linked to chromo夞 This work was supported by grants to L.I.B., N.G., and J.K. from the National Institutes of Health (AG09466; AG14452; GM56292) and the Alzheimer’s Association (RG2-96-076). * Corresponding author. Tel.: ⫹001-604-822-7379; fax: ⫹001-604822-7086. E-mail address: [email protected] (C. Schwab).

some 17 [12,13,27], including pallido-ponto-nigral degeneration (PPND), the disruption of the cytoskeleton can be directly linked to mutations in the tau gene [4,12,13,27]. In the more common “tauopathies” such as AD, other pathological processes lead to aberrant tau phosphorylation, accumulation of tau filaments, disruption of the cytoskeleton, and finally death of neurons. Efforts to identify phosphotransferases potentially involved in tau hyperphosphorylation have shown that multiple proline- and nonproline-directed protein kinases are potentially involved [15,19,20,30]. One of these enzymes, casein kinase I (CK1), fulfils several criteria expected of a pathological tau protein kinase. First, CK1 isoforms are capable of phosphorylating tau in vitro at sites found in filamentous-tau [34,35]. Moreover, all known CK1 isoforms have a “phosphate-directed” substrate selectivity, facilitating the progressive and hierarchical phosphorylation of substrate proteins such as tau to high stoichiometry [18,42]. Second, CK1 homologs colocalize with NFTs in

0197-4580/00/$ – see front matter © 2000 Elsevier Science Inc. All rights reserved. PII: S 0 1 9 7 - 4 5 8 0 ( 0 0 ) 0 0 1 1 0 - X

504

C. Schwab et al. / Neurobiology of Aging 21 (2000) 503–510

authentic AD tissue [8], placing them at what is thought to be a pathologically relevant site of action. Association with NFTs appears to be of high affinity, as CK1 isoforms remain associated with tau filaments through stringent purification methods [18]. Interestingly, CK1 isoforms also colocalize with granulovacuolar bodies (GVBs) in AD hippocampus [8]. These bodies are a second pathological hallmark of AD [1,41] that react strongly with monoclonal antibodies raised against phosphoepitopes [7,38]. GVBs appear in pyramidal and other large neurons in a variety of neurodegenerative diseases and, to a lesser extent, in aging normals. Finally, some CK1 isoforms are overexpressed in AD, with one isoform, Cki␦ being elevated greater than 30-fold in AD hippocampus relative to age-matched controls [8]. Together, these data support a role for CK1 in hyperphosphorylation of tau and potentially other proteins in AD. Here we extend the immunohistochemical analysis of Cki␦ to other diseases where pathological tau-containing lesions appear. The results show that Cki␦ colocalizes with NFTs in DS, PSP, PPND, and PDC, and with Pick bodies and ballooned neurons in PiD. However, it is not associated with the glial tau-containing inclusions in PSP, PPND, and PDC. Cki␦ is also not associated with tau-negative inclusions such as Lewy bodies in Parkinson’s disease (PD), and dementia with Lewy bodies (DLB), Hirano bodies, Marinesco bodies, and skein-like inclusions of amyotrophic lateral sclerosis (ALS).

Table 1 Description of cases studied

2. Materials and methods

7.6. When a dark blue color developed, sections were washed, mounted on glass slides, and coverslipped with Entellan. Controls performed by omitting the primary antibody were consistently negative for immunoreactivity. Selected sections were counterstained with neutral red. The criteria for distinguishing extracellular and intracellular NFTs [33], glial tangles [14,17,43], GVBs [8], and Hirano bodies [10] were as described previously.

Five cases of AD, four cases of PDC, three cases each of PSP, PiD, and PD, two cases each of DLB and PPND, one case of DS, and three age-matched control cases without known neurological symptoms were selected from the brain bank at the University of British Columbia (for case demographics see Table 1). Brain tissue was fixed in 4% paraformaldehyde and, after 3 to 4 days, transferred to a 15% buffered sucrose solution for long term storage. Immunohistochemistry was performed as previously reported [32]. Briefly, 30 ␮m sections were cut on a freezing microtome. The sections were then treated for 30 min with 0.3% H2O2 solution in 0.01M phosphate-buffered saline (PBS), pH 7.4, containing 0.3% Triton X-100 (PBS-T), transferred into 5% skim milk in PBS-T for 30 min, and incubated for 72 h at 4°C or overnight at room temperature with one of the primary antibodies listed in Table 2. Their source and dilutions are given in Table 2. Sections were next treated with appropriate biotinylated secondary antibodies for 2 h at room temperature, followed by incubation in avidin-biotinylated horseradish peroxidase complex (ABC Elite, Vector Lab, Burlingame, CA, USA) for 1 h at room temperature. Peroxidase labeling was visualized by incubation in 0.01% 3,3-diaminobenzidine (DAB, Sigma, St. Louis, MO, USA) containing 0.6% nickel ammonium sulfate and 0.001% H2O2 in 0.05 M Tris-HCl buffer, pH

Case

Age (years)

Sex

Postmortem delay (h)

AD 1 AD 2 AD 3 AD 4 AD 5 PDC 1 PDC 2 PDC 3 PDC 4 G-ALS ALS ALS PSP 1 PSP 2 PPND1 PPND2 PiD 1 PiD 2 PiD 3 DLB 1 DLB 2 PD 1 PD 2 PD 3 DS Co 1 Co 2 Co 3

58 82 68 71 78 60 77 67 67 61 65 81 62 60 48 63 64 73 65 73 63 75 81 77 50 78 89 85

f f m f f m m m f f f f m m f f f m f m m m m m m m f m

6 20 6 8 16.5 15 14 72 24 7 7 15 5 12 62 84 7 3.5 10 18 24 18 18 10 36 72 24 4

3. Results The results of immunostaining with the Cki␦ antibody are summarized in Table 3. As described previously, the Cki␦ antibody immunoreacted strongly with GVBs (granuTable 2 Source and dilution of antibodies Antibody

Antibody type

Cki␦

mouse IgG1

Antigen

Source

recombinant protein ICOS Corp. (6, 13) Cki␧ mouse IgG1 recombinant protein Transduction lab. tau-2 mouse IgG1 human tau Sigma (18) tau Rabbit polyclonal Chemicon Alz50 mouse IgM A68 Dr. Davies (29) Ubiquitin D Rabbit polyclonal ubiquitin Dako

Dilution 1:5000 1:1000 1:2000 1:2000 1:2000 1:1000

C. Schwab et al. / Neurobiology of Aging 21 (2000) 503–510 Table 3 Results of immunohistochemical staining Pathological entity Disease entity

Intensity of immunostaining Cki␦ Tau Ubiquitin D

GVB NFT

All studied cases AD, DS, PDC, PSP, PPND, Co DN in plaques AD, DS, PDC Neuropil threads AD, DS, PDC Pick bodies PiD Ballooned neurons PiD Glia tangles PDC, AD, PSP, PPND Lewy bodies PD, DLB Marinesco bodies AD, PDC, PD, Co Hirano bodies PDC Skein inclusions G-ALS AHC G-ALS, ALS

s v

— s

— s

v v s m — — — — — m

v s s m s — — — — m

v s v v s v s — v m

Abbreviations: GVB, granulovacuolar body; NFT, neurofibrillary tangle: DN, dystrophic neurites; AHC, anterior horn cells; AD, Alzheimer’s disease; DS, Down syndrome; PDC, Parkinsonism dementia complex of Guam; PSP, progressive supranuclear palsy; PPND, pallido-ponto-nigral degeneration; Co, control; PiD, Pick’s disease; PD, Parkinson’s disease; DLB, Lewy body disease; G-ALS, Guamanian amyotrophic lateral sclerosis; ALS, amyotrophic lateral sclerosis. Intensity of immunostaining: w weak, m moderate, s strong, v variable, — no staining.

lovacuolar bodies) in AD cortex and hippocampus (Fig. 1). NFTs were also stained with varying intensity. The stained NFTs were intracellular. Extracellular NFTs were stained only slightly above background (Fig. 1A). Cki␦ immunoreactivity was also associated with dystrophic neurites (DNs) and neuropil threads (Fig. 1D), especially in plaque areas. Overall, the hippocampus and entorhinal cortex were the brain areas most intensely stained, but some Cki␦ immunostaining of NFTs and GVBs was observed in the substantia nigra (SN) (Fig. 1E and F) and the nucleus basalis of Meynert. The patterns of immunostaining in DS and PDC was very similar to that in AD. Many GVBs, numerous NFTs, and some DNs were Cki␦ reactive in both conditions (Fig. 2A and D). In some cortical areas of PDC, the number of Cki␦ positive NFTs was even higher than in AD. In the aged controls, a few GVBs, and rare NFTs were observed. More NFTs were immunoreactive with the tau antibodies than the antibody to Cki␦. GVBs were not stained by tau-2 (Fig. 2A and B). The proportion of NFTs stained by Cki␦ varied between cases and regions. In CA1 of three of the AD cases, the number of Cki␦ positive NFTs was 40%, 46%, and 47% of the number of tau positive NFTs. The percentage was high in CA4 of one AD case (71%), but in another AD case there were no Cki␦ positive NFTs in this region. In the PPND cases, tau immunoreactive inclusions were found in neurons and oligodendroglia (Fig. 3A and C) as reported previously [28,43]. Many Cki␦ positive GVB and some Cki␦ positive NFTs were found in the areas with tau immunoreactive neuronal tangles, as shown in Fig. 3B. But no Cki␦ immunoreactivity was associated with the tau pos-

505

itive inclusions of oligodendroglia cells (coiled bodies, Fig. 3C and D). Likewise, tau-containing astroglia inclusions in PDC (Fig. 4) did not immunostain with Cki␦. Cki␦ immunoreactivity was particularly strongly associated with the pathological lesions in PiD. The numbers of Cki␦-immunoreactive Pick bodies (PB) and ballooned neurons were similar to the numbers observed in immunostaining for the polyclonal tau antibody (Fig. 5A-C). Additionally, GVBs were stained intensely in these cases. The tau-2 antibody stained ballooned neurons strongly and Pick bodies occasionally. Pick bodies were stained also occasionally and with variable intensity with the antibody against ubiquitin. As in AD and control cases, GVBs were tau-2 negative. In the basal ganglia, amygdala and cortex of the PSP cases, a few NFTs, as well as some GVBs, contained Cki␦ immunoreactivity (Fig. 2C). The number of anterior horn motor neurons was reduced in the cases with ALS and Guamanian ALS (lytico). In Guamanian ALS, the remaining neurons had skein-like inclusions. These inclusions stained with an antibody against ubiquitin, but not with the Cki␦ or tau-2 antibodies. However, the cytoplasm of these motor neurons contained Cki␦ positive granules (Fig. 5D). The granules were smaller and more diffuse than the GVBs in the cortex and hippocampus. The same granules were found to be stained by the tau-2 antibody. In classic ALS cases, no skein-like inclusions were found, but some motor neurons of the spinal cord showed similar Cki␦ and tau-2 immunoreactive granules. These Cki␦ positive granules in ALS motoneurons are most likely different from lipofuscin inclusions. That is apparent in Fig. 5D, where in the strongly stained neuron the lipofuscin occupies an unstained area left of the nucleus. Lewy bodies (LB) in PD and DLB were generally not stained by Cki␦ (Fig. 4B). Marinesco and Hirano bodies, as well as tau-immunoreactive astroglial tangles, were negative for Cki␦ immunoreactivity (Fig. 4C and D). The antibody against Cki⑀ demonstrated a similar pattern of immunoreactivity to that for Cki␦, although the staining was less intense (Fig. 4A).

4. Discussion We have shown that Cki␦ immunoreactivity colocalizes with tau-bearing lesions found in a range of neurodegenerative disorders. These include the neuronal NFTs of AD, DS, PDC, PPND, and PSP, and the ballooned neurons and Pick bodies of Pick’s disease. All of these lesions contain abnormally phosphorylated forms of tau, although their molecular structure and anatomical distribution varies. The pathological forms of tau from AD, DS, and PDC demonstrate four bands on Western blots (72, 68, 64, and 60 kDa; Type I pattern) [36]. All six isoforms of tau are represented in the NFTs [9]. Pathological tau from Pick’s disease demonstrates two bands (64 and 60 kDa; Type II pattern). Only three of the isoforms are represented, with an apparent

506

C. Schwab et al. / Neurobiology of Aging 21 (2000) 503–510

Fig. 1. Cki␦ immunostaining in Alzheimer’s disease. A, many NFTs and GVBs are immunoreactive in the hippocampal CA2 region. Note the numerous extracellular NFTs, which are weakly gray and not stained by the Cki␦ antibody. B, occasionally Cki␦ immunoreactive GVBs and NFTs occur in the same neuron (arrows). C, Cki␦ immunoreactive GVBs in the perikaryon and proximal dendrites of CA1 pyramidal neurons. D, dystrophic neurites in plaques next to GVBs in neurons of the CA4 regions (arrows). E, F, Cki␦ immunoreactive NFTs and GVBs in substantia nigra neurons. Calibration bars in A, for A, C, and D, 100 ␮m; in B, for B, E, and F, 50 ␮m.

deficiency of the three isoforms with four repeat binding sequences [6]. Pathological tau from PSP demonstrates three bands (72, 68, and 64 kDa; Type III pattern) [36]. Again, only three isoforms are represented, but these are the three that incorporate the four tandem repeats. In contrast to these lesions, all of which contain multiple forms of abnormally phosphorylated tau, Cki␦ did not colocalize with tau-negative lesions such as LBs, Marinesco bodies, Hirano bodies, or the skein-like inclusions of ALS. Together these results suggest that neurons accumulating filamentous tau inclusions, regardless of the isoforms of tau or the location of the neuronal population, also accumulate elevated levels of Cki␦. In contrast, tau immunoreactive glial tangles in PDC (astroglia) and PPND (oligodendroglial coiled bodies) were not immunostained for Cki␦. Likewise, glial tangles of PSP, which also contain phosphorylated forms of tau, but not tau filaments [14], do not immunostain for Cki␦. The absence of classical tau filaments in these lesions may be related to the absence of Cki␦. As previously noted [8], Cki␦ is primarily a neuronal enzyme. Thus, the failure to detect this isoform

of CK1 in glial tangles may be the result of poor glial expression. It is noteworthy that the glial tangles of PSP also fail to stain for several other proteins that associate with neuronal NFTs in AD, DS, and PDC. These include ubiquitin, apoE, ␣1-antichymotrypsin, and heparan sulfate [14]. In the case of apoE, the failure is unlikely to be due to deficient expression in astrocytes, as they are known to be a significant source of brain apoE [26]. Other undetermined factors may play a role in the different associations between abnormally phosphorylated tau and other proteins, including Cki␦, in glial as compared with neuronal tangles. The results presented here also confirm the remarkable association of Cki␦ with AD GVBs [8] and extend the analysis to include these lesions as they appear in DS, PiD, PDC, and PSP. As before [8], GVBs were not stained by antibodies against nonphosphorylated tau protein (tau-2, Alz50, tau – Chemicon, Temecula, CA, USA). There are differing reports on tau immunoreactivity of GVBs, and it has been suggested that these bodies may contain tau sequestered in a conformation that is inaccessible to some antibodies [2]. Others have sug-

C. Schwab et al. / Neurobiology of Aging 21 (2000) 503–510

507

Fig. 2. A, NFTs and GVBs are Cki␦ immunoreactive in PDC CA1. B, in sequential sections, immunoreacted with the tau2 antibody, only NFTs are stained. C, a few NFTs are stained for Cki␦ in the amygdala of a PSP case. D, intense Cki␦ immunoreactivity in the subiculum of a Down syndrome case. GVBs and dystrophic neurites of plaques are stained. Calibration bar in D for A-D: 100 ␮m.

gested that only a portion of tau may be present in these bodies [7]. For example, Dickson et al. [7] found that GVBs did not stain with antibodies against nonphospho-

rylated tau exon 2, or against the amino-terminal and carboxy-terminal and certain mid regions. An antibody against exon 3, however, resulted in consistent staining

Fig. 3. In PPND, neuronal inclusions (NFTs) in the substantia innominata show tau (A) and Cki␦ (B) immunoreactivity. Oligodendroglia tau-containing inclusions (coiled bodies) in the commissura anterior (C) are not Cki␦ immunoreactive (D). Calibration bar in D for A-D, 50 ␮m.

508

C. Schwab et al. / Neurobiology of Aging 21 (2000) 503–510

Fig. 4. A, Cki⑀ immunoreactivity shows a similar distribution to Cki␦ as shown in the CA1 region of an AD case. B, Lewy bodies in the substantia nigra in Parkinson’s disease are not stained by the Cki␦ antibody (arrows). C, Tau2 immunoreactive astroglial tangles in the hilar region of PDC are not Cki␦ immunoreactive, as demonstrated in a parallel section (D). Calibration bars in D for A, C, D, 100 ␮m; in B, 50 ␮m.

of GVBs. Dickson et al. [7] suggested that GVBs contain the amino half of tau without the amino terminus. Although truncation may account for the failure of Alz50 to recognize putative tau in GVBs [3], it does not explain the behavior of Tau-2, which binds residues Ala [106] Lys [130] of htau40 [40]. Stadelmann et al. [37] demonstrated recently a possible association between tau and GVB: neurons containing aberrantly phosphorylated tau (pretangled or tangled neurons) are more likely to contain GVBs. Despite these conflicting data, it is clear that GVBs cross react strongly with monoclonal antibodies raised against various phosphoproteins [7,38], suggesting that these lesions, like the tau filaments of NFTs, are rich in covalently bound phosphate. Because of its spatial distribution and catalytic activity, Cki␦ is positioned to play a role in creating the phosphoepitopes detected in GVBs by various phosphorylation-dependent monoclonal antibodies. GVBs are thought to be a product of autophagy, a degradative process that functions by a nonselective volume uptake mechanism, and that uses a modification cascade distinct from the ubiquitin pathway [21,22]. Using Cki␦ as a robust marker for these lesions confirms that the process of autophagy occurs in widespread brain areas, although it is most prominent in the hippocampus. On the basis of immunostaining, the results presented here suggest that a

similar process is also underway in anterior horn motor neurons of ALS. Thus, Cki␦ is a marker for a degradative process common to degenerating neurons in diverse regions of the central nervous system. Once considered a single entity, human CK1 is now known to consist of multiple isoforms encoded by distinct genes (Cki ␣,␥1,␥2,␥3,␦,⑀). Although their function is not understood in molecular detail, recent evidence suggests they may play a role in regulation of DNA-repair [11], cellular morphology [29], circadian rhythm [16], and stabilization of cellular proteins such as ␤-catenin [25,31] and membrane-bound transporters [5] against degradation. These phenotypes, combined with observations that different CK1 homologs are essential for intracellular trafficking in different cellular compartments [23,24,39], suggest that CK1 isozymes directly influence protein turnover and other transport-dependent cellular processes such as autophagy, secretion, and phagocytosis. Appearance of one CK1 isoform, Cki␦ in tau-containing neuronal inclusions (NFTs, Pick bodies, and ballooned neurons), but not in tau negative inclusions (Lewy bodies, Hirano bodies, and Marinesco bodies), is consistent with this isoform participating in the pathological hyperphosphorylation of tau protein. In vitro, tau can be phosphorylated to high stoichiometry by CK1 at multiple sites, including those found occupied in authentic PHF tau (e.g., Ser396 and

C. Schwab et al. / Neurobiology of Aging 21 (2000) 503–510

509

Fig. 5. In Pick disease, Cki␦ immunoreactivity was found in Pick bodies and GVBs. A, low-power microphotograph of the hippocampus with many Pick bodies in CA1 pyramidal neurons. B, ballooned neurons of Pick disease contain Cki␦ immunoreactive GVBs, as shown here in the midtemporal cortex. C, higher magnification of CA1 neurons. Pick bodies and GVBs are immunostained. D, an anterior horn motor neuron in ALS with Cki␦ immunoreactive granules. Calibration bars in A, 500 ␮m; in C, 100 ␮m; and in D, for B and D, 50 ␮m.

Ser404; [34]). As discussed previously, the phosphate-directed substrate selectivity of CK1 is appropriate for amplifying tau phosphorylation and may contribute to the high phosphorylation stoichiometries observed in authentic ADderived PHF-tau [18]. In summary, we have shown that Cki␦ colocalizes with neurofibrillary and granulovacuolar lesions in a variety of neurodegenerative diseases. Clarifying the mechanism of Cki␦ accumulation in disease will yield insight into the pathway of neurodegeneration common to AD, PSP, and ALS.

Acknowledgments We thank Dr P. Davies for his gift of the Alz50 antibody and Joane Sunahara for technical assistance. This work was supported by grants from the Parkinson Foundation of Canada, the Alzheimer Society of British Columbia, and the Jack Brown and Family A.D. Research Fund, as well as donations from individual British Columbians.

References [1] Ball MJ. Topographic distribution of neurofibrillary tangles and granulovacuolar degeneration in hippocampal cortex of aging and demented patients. Acta Neuropathol 1978;42:73– 80. [2] Bondareff W, Wischik CM, Novak M, Roth M. Sequestration of tau by granulovacuolar degeneration in Alzheimer’s disease. Am J Pathol 1991;139:641–7. [3] Carmel G, Mager EM, Binder LI, Kuret J. The structural basis of monoclonal antibody Alz50’s selectivity for Alzheimer’s Disease pathology. J Biol Chem 1996;271:32789 –95. [4] Clark LN, Poorkaj P, Wszolek Z, Geschwind DH, Nasreddine ZS, Miller B, Li D, Payami H, Awert F, Markopoulou K, Andreadis A, D’Souza I, Lee VM, Reed L, Trojanowski JQ, Zhukareva V, Bird T, Schellenberg G, Wilhelmsen KC. Pathogenic implications of mutations in the tau gene in pallido-ponto-nigral degeneration and related neurodegenerative disorders linked to chromosome 17. Proc Natl Acad Sci USA 1998;95:13103–7. [5] Decottignies A, Owsianik G, Ghislain M. Casein kinase I-dependent phosphorylation and stability of the yeast multidrug transporter Pdr5p. J Biol Chem 1999;274:37139 – 46. [6] Delacourte A, Sergeant N, Wattez A, Gauvreau D, Robitaille Y. Vulnerable neuronal subsets in Alzheimer’s and Pick’s disease are distinguished by their tau isoform distribution and phosphorylation. Ann Neurol 1998;43:193–204.

510

C. Schwab et al. / Neurobiology of Aging 21 (2000) 503–510

[7] Dickson DW, Liu WK, Kress Y, Ku J, DeJesus O, Yen SH. Phosphorylated tau immunoreactivity of granulovacuolar bodies (GVB) of Alzheimer’s disease: localization of two amino terminal tau epitopes in GVB. Acta Neuropathol 1993;85:463–70. [8] Ghoshal N, Smiley JF, DeMaggio AJ, Hoekstra MF, Cochran EJ, Binder LI, Kuret J. A new molecular link between the fibrillar and granulovacuolar lesions of Alzheimer’s disease. Am J Pathol 1999; 155:1163–72. [9] Goedert M, Spillantini MG, Davies SW. Filamentous nerve cell inclusions in neurodegenerative diseases. Curr Opin Neurobiol 1998; 8:619 –32. [10] Hirano A. Hirano bodies and related neuronal inclusions. Neuropathol Appl Neurobiol 1994;20:3–11. [11] Hoekstra MF, Liskay RM, Ou AC, DeMaggio AJ, Burbee DG, Heffron F. HRR25, a putative protein kinase from budding yeast: association with repair of damaged DNA. Science 1991;253:1031– 4. [12] Hong M, Zhukareva V, Vogelsberg RV, Wszolek Z, Reed L, Miller BI, Geschwind DH, Bird TD, McKeel D, Goate A, Morris JC, Wilhelmsen KC, Schellenberg GD, Trojanowski JQ, Lee VM. Mutation-specific functional impairments in distinct tau isoforms of hereditary FTDP-17. Science 1998;282:1914 –7. [13] Hutton M, Lendon CL, Rizzu P, Baker M, Froelich S, Houlden H, Pickering BS, Chakraverty S, Isaacs A, Grover A, Hackett J, Adamson J, Lincoln S, Dickson D, Davies P, Petersen RC, Stevens M de GE, Wauters E van BJ, Hillebrand M, Joosse M, Kwon JM, Nowotny P, Heutink P, et al. Association of missense and 5⬘-splice-site mutations in tau with the inherited dementia FTDP-17. Nature 1998;393: 702–5. [14] Ikeda K, Akiyama H, Arai T, Nishimura T. Glial tau pathology in neurodegenerative diseases: their nature and comparison with neuronal tangles. Neurobiol Aging 1998;19:S85–91. [15] Imahori K, Hoshi M, Ishiguro K, Sato K, Takahashi M, Shiurba R, Yamaguchi H, Takashima A, Uchida T. Possible role of tau protein kinases in pathogenesis of Alzheimer’s disease. Neurobiol Aging 1998;19:S93– 8. [16] Kloss B, Price JL, Saez L, Blau J, Rothenfluh A, Wesley CS, Young MW. The Drosophila clock gene double-time encodes a protein closely related to human casein kinase I epsilon. Cell 1998;94:97– 107. [17] Komori T. Tau-positive glial inclusions in progessive supranuclear palsy, corticobasal degeneration and Pick’s disease. Brain Pathol 1999;9:663–79. [18] Kuret J, Johnson GS, Cha D, Christenson ER, DeMaggio AJ, Hoekstra MF. Casein kinase 1 is tightly associated with paired-helical filaments isolated from Alzheimer’s disease brain. J Neurochem 1997;69:2506 –15. [19] Masliah E, Cole GM, Hansen LA, Mallory M, Albright T, Terry RD, Saitoh T. Protein kinase C alteration is an early biochemical marker in Alzheimer’s disease. J Neurosci 1991;11:2759 – 67. [20] Masliah E, Iimoto DS, Mallory M, Albright T, Hansen L, Saitoh T. Casein kinase II alteration precedes tau accumulation in tangle formation. Am J Pathol 1992;140:263– 8. [21] Mizushima N NT Yoshimori T, Tanaka Y, Ishii T, George MD, Kionsky DJ, Ohsumi M, Oshumi Y. A protein conjugation system essential for autophagy. Nature 1998;395:395– 8. [22] Mortimore GE, Miotto G, Venerando R, Kadowaki M. Autophagy. Subcell Biochem 1996;27:93–135. [23] Murakami A, Kimura K, Nakano A. The inactive form of a yeast casein kinase I suppresses the secretory defect of the sec12 mutant. Implication of negative regulation by the Hrr25 kinase in the vesicle budding from the endoplasmic reticulum. J Biol Chem 1999;274: 3804 –10. [24] Panek HR, Stepp JD, Engle HM, Marks KM, Tan PK, Lemmon SK, Robinson LC. Suppressors of YCK-encoded yeast casein kinase 1 deficiency define the four subunits of a novel clathrin AP-like complex. EMBO J 1997;16:4194 –204.

[25] Peters JM, McKay RM, McKay JP, Graff JM. Casein kinase I transduces Wnt signals. Nature 1999;401:345–50. [26] Pitas RE, Boyles JK, Lee SH, Foss D, Mahley RW. Astrocytes synthesize apolipoprotein E and metabolize apolipoprotein E-containing lipoproteins. Biochim Biophys Acta 1987;917:148 – 61. [27] Poorkaj P, Bird TD, Wijsman E, Nemens E, Garruto RM, Anderson L, Andreadis A, Wiederholt WC, Raskind M, Schellenberg GD. Tau is a candidate gene for chromosome 17 frontotemporal dementia. Ann Neurol 1998;43:815–25. [28] Reed LA, Schmidt ML, Wszolek ZK, Balin BJ, Soontornniyomkij V, Lee VM, Trojanowski JQ, Schelper RL. The neuropathology of a chromosome 17-linked autosomal dominant parkinsonism and dementia (“pallido-ponto-nigral degeneration”). J Neuropathol Exp Neurol 1998;57:588 – 601. [29] Robinson LC, Menold MM, Garrett S, Culbertson MR. Casein kinase I-like protein kinases encoded by YCK1 and YCK2 are required for yeast morphogenesis. Mol Cell Biol 1993;13:2870 – 81. [30] Saitoh T, Masliah E, Jin LW, Cole GM, Wieloch T, Shapiro IP. Protein kinases and phosphorylation in neurologic disorders and cell death. Lab Invest 1991;64:596 – 616. [31] Sakanaka C, Leong P, Xu L, Harrison SD, Williams LT. Casein kinase-I epsilon in the wnt pathway: regulation of beta-catenin function. Proc Natl Acad Sci USA 1999;96:12548 –52. [32] Schwab C, Akiyama H, McGeer EG, McGeer PL. Extracellular neurofibrillary tangles are immunopositive for the 40 carboxy-terminal sequence of §-amyloid protein. J Neuropath Exp Neurol 1998;57: 1131–7. [33] Schwab C, Schulzer M, Steele JC, McGeer PL. On the survival time of a tangled neuron in the hippocampal CA4 region in Parkinson dementia complex of Guam. Neurobiol Aging 1999;20:57– 63. [34] Singh TJ, Grundke II, Iqbal K. Phosphorylation of tau protein by casein kinase-1 converts it to an abnormal Alzheimer-like state. J Neurochem 1995;64:1420 –3. [35] Singh TJ, Grundke II, McDonald B, Iqbal K. Comparison of the phosphorylation of microtubule-associated protein tau by non-proline dependent protein kinases. Mol Cell Bioch 1994;131:181–9. [36] Spillantini MG, Goedert M. Tau protein pathology in neurodegenerative diseases. Trends Neurosci 1998;21:428 –33. [37] Stadelmann C, Deckwerth TL, Srinivasan A, Bancher C, Bru¨ck W, Jellinger K, Lassmann H. Activation of caspase-3 in single neurons and autophagic granules of granulovacuolar degeneration in Alzheimer disease. Am J Pathol 1999;155:1459 – 66. [38] Vincent I, Zheng J-H, Dickson DW, Kress Y, Davies P. Mitotic phosphoepitopes precede paired helical filaments in Alzheimer’s Disease. Neurobiol Aging 1998;19:287–96. [39] Wang X, Hoekstra MF, DeMaggio AJ, Dhillon N, Vancura A, Kuret J, Johnston GC, Singer RA. Prenylated isoforms of yeast casein kinase I, including the novel Yck3p, suppress the gcs1 blockage of cell proliferation from stationary phase. Mol Cell Biol 1996;16:5375– 85. [40] Watanabe N, Takio K, Hasegawa M, Arai T, Titani K, Ihara Y. Tau 2: a probe for a Ser conformation in the amino terminus of tau. J Neurochem 1992;58:960 – 6. [41] Xu M, Shibayama H, Kobayashi H, Yamada K, Ishihara R, Zhao P, Takeuchi T, Yoshida K, Inagaki T, Nokura K. Granulovacuolar degeneration in the hippocampal cortex of aging and demented patients—a quantitative study. Acta Neuropath 1992;85:1–9. [42] Xu R, Carmel G, Sweet RM, Kuret J, Cheng X. Crystal structure of casein kinase-1, a phosphate directed protein kinase. EMBO J 1995; 14:1015–23. [43] Yamada T, McGeer PL, McGeer EG. Appearance of paired nucleated, tau-positive glia in patients with progressive supranuclear palsy brain tissue. Neurosci Let 1992;135:99 –102.