DJ-1 (PARK7) is associated with 3R and 4R tau neuronal and glial inclusions in neurodegenerative disorders

www.elsevier.com/locate/ynbdi Neurobiology of Disease 28 (2007) 122 – 132

DJ-1 (PARK7) is associated with 3R and 4R tau neuronal and glial inclusions in neurodegenerative disorders Ravindran Kumaran,a Ann Kingsbury,a Ian Coulter,a Tammaryn Lashley,b David Williams,a Rohan de Silva,a David Mann,c Tamas Revesz,b Andrew Lees,a and Rina Bandopadhyay a,⁎ a

Reta Lila Weston Institute of Neurological Studies, Institute of Neurology, 1, Wakefield Street, WC1N 1PJ, UK Department of Molecular Neuroscience and Queen Square Brain Bank, Institute of Neurology, Queen Square, WCIN 3BG, UK c Clinical Neurosciences Research Group, University of Manchester, Greater Manchester Neurosciences Centre, Hope Hospital, Salford, M6 8HD, UK b

Received 24 April 2007; revised 25 June 2007; accepted 1 July 2007 Available online 18 July 2007 Mutations in the DJ-1 gene are associated with autosomal recessive Parkinson’s disease (PD), but its role in disease pathogenesis is unknown. This study examines DJ-1 immunoreactivity (DJ-1 IR) in a variety of neurodegenerative disorders, Alzheimer’s disease (AD), frontotemporal lobar degeneration (FTLD) with Pick bodies, FTLD with MAPT mutations, progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD), in which hyperphosphorylated tau inclusions are the major pathological signature. DJ-1 IR was seen in a subset of neurofibrillary tangles (NFTs), neuropil threads (NTs), and neurites in extracellular plaques in AD; tau inclusions in AD contained both 3R and 4R tau. A subset of Pick bodies in FTLD showed DJ-1 IR. In PSP, DJ-1 IR was present in a few NFTs, NTs and glial cell inclusions. In CBD, DJ-1 IR was seen only in astrocytic plaques. In cases of FTLD with MAPT mutations that were 4R tau positive (i.e. N279K and exon 10 + 16 mutations), DJ-1 IR was present mostly in oligodendroglial coiled bodies. However, in MAPT R406W mutation cases, DJ-1 IR was associated mainly with NFTs and NTs and these were both 3R and 4R tau positive. No DJ-1 IR was present in FTLD with ubiquitin inclusions (FTLD-U). In AD and FTLD with Pick bodies, DJ-1 protein was enriched in the sarkosyl-insoluble fractions of frozen brain tissue containing insoluble hyperphosphorylated tau, thus strengthening the association of DJ-1 with tau pathology. Additionally using twodimensional gel electrophoresis, we demonstrated accumulation of acidic pI isoforms of DJ-1 in AD brain, which may compromise its normal function. Our observations confirm previous findings that DJ-1 is present in a subpopulation of glial and neuronal tau inclusions in tau diseases and associated with both 3R and 4R tau isoforms. © 2007 Elsevier Inc. All rights reserved. Keywords: DJ-1; Tauopathies; Hyperphosphorylated tau; 3R tau; 4R tau

⁎ Corresponding author. E-mail address: [email protected] (R. Bandopadhyay). Available online on ScienceDirect (www.sciencedirect.com). 0969-9961/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.nbd.2007.07.012

Introduction Recently homozygous mutations in the DJ-1 gene (DJ-1) were identified in two families linked to autosomal recessive Parkinson’s disease (PD); a deletion spanning the first five exons, leading to an absence of DJ-1 protein, was present in the Netherlands family, whilst a homozygous point mutation (L166P) was shown to cause parkinsonian features in an Italian family (Bonifati et al., 2003). The latter point mutation destabilises the protein leading to its enhanced degradation by the proteasome (Miller et al., 2003). Further separate mutations in DJ-1 have since been reported in PD patients, including one in PD-ALS complex of Guam (AbouSleiman et al., 2003; Annesi et al., 2005; Hague et al., 2003; Hedrich et al., 2004). No patient bearing a homozygous DJ-1 mutation has yet come to post mortem and therefore it is not clear whether affected patients harbour alpha-synuclein-positive Lewy bodies (LBs), the pathological hallmark typical of sporadic PD. Loss of DJ-1 function leads to parkinsonian features but the molecular mechanisms are yet to be elucidated. DJ-1 was originally described as an oncogene (Nagakubo et al., 1997) and in parallel was found to encode a protein involved in male fertility in rats (Wagenfeld et al., 1998). Transgenic animal experiments have shown that disruption of DJ-1 expression in mice leads to altered dopamine D2 receptor signalling (Chen et al., 2005; Goldberg et al., 2005) and increased susceptibility to MPTP (Kim et al., 2005a). Furthermore, DJ-1 deficient dopaminergic neurons showed an increased sensitivity to oxidative stress (Martinat et al., 2004) and nigral neurons from DJ-1 KO mice demonstrated an increased susceptibility to energy deprivation (Pisani et al., 2006). DJ-1 protein comprises 189 amino acids and is ubiquitously expressed. It forms soluble dimers and has been predicted to have multiple putative functions (Tao and Tong, 2003; Wilson et al., 2003). DJ-1 also works as a regulatory subunit of an RNA-binding complex (Hod et al., 1999) and has structural homology to a bacterial protease, an E coli chaperone protein Hsp31 and the YajL/ ThiJ group (Huai et al., 2003; Lee et al., 2003; Wei et al., 2007; Bandyopadhyay and Cookson, 2004). Indeed DJ-1 protein has

R. Kumaran et al. / Neurobiology of Disease 28 (2007) 122–132

been shown to act as a chaperone and to participate in oxidative stress responses in vitro (Lee et al., 2003; Mitsumoto and Nakagawa, 2001; Shendelman et al., 2004; Taira et al., 2004). Our previous work examining DJ-1 protein expression in the cerebral cortex and substantia nigra of control and PD brain demonstrated that DJ-1 was mostly localised to astrocytes, with very little protein expression in neurones. In addition, we have demonstrated an accumulation of acidic pI isoforms of DJ-1 protein in PD brain, suggesting that the normal function of DJ-1 may be compromised in sporadic PD (Bandopadhyay et al., 2004). Two previous studies have linked the DJ-1 protein with other neurological diseases like Alzheimer’s disease (AD) and frontotemporal lobar degeneration (FTLD) by virtue of its co-existence with tau inclusions, suggesting that DJ-1 may be involved in the pathogenesis of diverse neurological disorders (Neumann et al., 2004; Rizzu et al., 2004). Neither of these studies, however, correlated the presence of DJ-1 with the different isoforms of tau. Increasing evidence suggests that synucleinopathies and tauopathies may share common pathogenic mechanisms (Galpern and Lang, 2006). The precise role of DJ-1 in the human brain has yet to be elucidated. Therefore, unravelling the physiological role of DJ-1 in the human brain in health and disease may reveal functions that are neuroprotective to susceptible cell types (e.g. dopaminergic neurons) and whose perturbation may be central to the mechanisms underlying neurodegeneration in a number of diseases. Hyperphosporylated tau accumulation in neurofibrillary tangles (NFTs), neuropil threads (NTs) and glial cells are characteristic features of a number of tauopathies like AD, FTLD, FTLD-MAPT (with pathogenic mutations in tau), PSP and CBD. Alternative splicing of microtubule-associated protein tau coded by a single gene on chromosome 17q21 produces tau isoforms with either three (3R) or four (4R) repeat domains (reviewed in Goedert, 2005). The exact mechanisms by which tau gets phosphorylated followed by misfolding and accumulation in neurons and glia remain obscure. To further our understanding of the function of DJ-1 in the human brain, and its association with pathological tau in tauopathies, we have performed a detailed investigation of the expression of DJ-1 and its co-localisation with 3R and 4R tau isoforms using immunohistochemistry (IH) and double fluorescence confocal-microscopy (DFCM). Diseases studied were: AD, sporadic FTLD with Pick bodies and inherited FTLD with MAPT mutations, FTLD with ubiquitin pathology (FTLD-U), progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD). The molecular composition of DJ-1 was assessed in sarkosylinsoluble fractions enriched in paired helical filaments of tau in AD and Pick bodies in FTLD, and compared to that of controls. Additionally the pI isoforms of DJ-1 protein were analysed using 2 dimensional gel electrophoresis (2DGE). Materials and methods Tissues A total of 44 post mortem brains were studied. Brain tissues from cases #1–31 (Table 1) were obtained from the brain tissue archive at Queen Square Brain Bank following ethical review by the London Multicentre Research Ethics Committee and the National Hospital for Neurology and Neurosurgery/Institute of Neurology Joint Research Ethics Committee. Formalin-fixed brain tissues from cases #32–39 and #42–44 (Table 1) were obtained with Ethical consent from the Manchester Brain Bank. For cases

123

#40 and 41, only paraffin sections were available, these being kindly donated by Dr R Ravid (Netherlands Brain Bank, cases #42 and 43) and Dr. Zbigniew Wsozlek (Mayo Clinic, Jacksonville). Case details are presented in Table 1. Flash or slow frozen brain tissues were available from 3 controls (#1, 5 and 7), 3 AD cases (#11–13), and 3 sporadic FTLD cases with Pick bodies (#37–39) (Table 1) and these were investigated by biochemistry. Specific regions examined for each disease are defined in Table 2. Immunohistochemistry Immunohistochemistry (IH) was performed on 8 μm formalinfixed wax-embedded sections of frontal cortex, temporal cortex and hippocampus, basal ganglia and pons (where appropriate or available). For DJ-1 immunoreaction (DJ-1 IR), sections were dewaxed in xylene and treated with 95% formic acid for 5 min. Endogenous peroxidase reactions were blocked by treatment with 0.3% hydrogen peroxide in methanol for 10 min at room temperature, and sections were subsequently incubated with 5% milk (Marvel) to block non-specific reactions. Sections were incubated overnight at 4 °C in polyclonal DJ-1 antibody, 1130 (a generous gift of P. Rizzu and P. Heutink, University of Rotterdam) diluted 1:100 followed by “ready to use” secondary antibody (Broad Spectrum, Kit no: 85-9643, Zymed, USA) and streptavidin–peroxidase conjugate (Zymed, USA). DJ-1 IR was visualised with diaminobenzidine. Sections were counterstained lightly with haematoxylin. For immunohistochemistry using AT8 at 1:500 (Innogenetics, UK), 3R (RD3) at 1:3000, 4R (RD4) at 1:100 tau monoclonal antibodies and polyclonal anti-ubiquitin antibody at 1:100 (Dako UK), sections were pre-treated by pressure cooking in citrate buffer (pH 6.0) for 10 min to expose antigenic sites as previously described (de Silva et al., 2003). Immunofluorescence confocal microscopy Double immunofluorescence was carried out using DJ-1 and AT8 antibodies sequentially. DJ-1 signal was visualised using tetramethyl rhodamine labelled secondary antibody and AT8 with the fluorescein signal amplification kit (Perkin Elmer, UK). Sections were washed thoroughly in PBS and mounted in aquamount (Merck, UK). Control sections incubated without primary antibody exhibited no significant background staining. Fluorescent signals from sections were scanned using a Leica TCS40 laser confocal microscope. Semi-quantitative assessment of pathological inclusions The number of DJ-1 IR inclusions was compared with that of AT8-positive tau inclusions from similar sections of each region. Two researchers performed the semiquantitative assessments independently. Results are categorised in the range as absent (−) to numerous (++++) (Table 2). Biochemical analysis Soluble DJ-1 protein levels Flash or slow frozen samples (1 g) of frontal cortex from 3 control cases (# 1, 5 and 7), 3 AD cases (# 11–13) and 3 FTLD cases with Pick bodies (#37–39) were homogenised mechanically in 10 volumes of buffer containing 50 mM Tris–HCl buffer pH 7.0, 8 M NaCl, 1 mM EGTA, 10% sucrose and protease inhibitors

124

R. Kumaran et al. / Neurobiology of Disease 28 (2007) 122–132

Table 1 Demographic data of brain cases Case no.

Diagnosis

Gender

Age at death (years)

Disease duration (years)

Brain weight (g)

PMD (h)

Main pathology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

Control a Control Control Control Controla Control Controla AD AD AD ADa ADa ADa CBD CBD CBD CBD FTLD-U FTLD-U FTLD-U FTLD-U PSP PSP PSP PSP PSP PSP PSP PSP PSP PSP FTLD-Pick's FTLD-Pick's FTLD-Pick's FTLD-Pick's FTLD-Pick's FTLD-Pick'sa FTLD-Pick'sa FTLD-Pick'sa

F M F F M M F F M F F F M M M M M F M F M M F M F F M M M M M F M F M M F M F

83 86 77 81 81 67 83 92 64 91 94 57 82 80 64 69 65 78 69 55 70 90 78 62 63 90 78 72 69 80 74 69 56 62 74 68 60 61 50

– – – – – – – 7 na na na 11 10 5 4 na na 4 5 5 3 23 9 7 2 33 6 21 6 24 13 10 10 10 11 8 7 14 8

1360 1300 1250 1260 1185 1469 1350 1400 1055 1235 1160 1200 1024 1375 1199 1441 1176 1153 1119 1300 1235 1340 1407 1140 1225 1235 1222 1274 1139 1150 1449 850 1150 928 990 895 960 980 1065

20 53 23 14 40 22 49 24 64 na 28 26 38 66 42 7 24 37 28 44 24 36 33 na 51 37 66 27 44 70 37 48 26 46 23 31 24 36 26

Minor age related NFT in hippocampus Normal with some vascular pathology Normal with some vascular pathology Normal with some vascular pathology Minor age related NFT in hippocampus Normal with some vascular pathology Normal with some vascular pathology AD pathology Braak Stage VI AD pathology Braak Stage V AD pathology Braak Stage VI AD pathology NFT, massive tau accumulation AD pathology NFT, NT, CB NFT, NT, gliosis Numerous astrocytic plaques, mild NFT NFT, NT, glial inclusions Ubiquitin inclusions Ubiquitin inclusions Ubiquitin inclusions Ubiquitin inclusions NFT, NT, TA, CB NFT, NT, TA, CB NFT, NT, TA, CB NFT, NT, TA, CB NFT, NT, TA, CB NFT, NT, TA, CB NFT, NT, TA, CB NFT, NT, TA, CB NFT, NT, TA, CB NFT, NT, TA, CB PBs PBs PBs PBs PBs PBs PBs PBs

FTLD-MAPT 40 41 42 43 44

R406W R406W N279K Exon 10 + 16 Exon 10 + 16

M F F M F

70 71 53 55 65

7 13 8 5 13

1121 905 1000 1240 1040

na na na 34 23

NFT NFT NFT, glial cell tangles NFT, glial cell tangles NFT, glial cell tangles

a

Flash/slow frozen cases for biochemical analysis and na = not available.

(Roche, UK). The homogenates were spun at 5000 rpm to remove debris. Protein concentrations from supernatants were determined using Bio-Rad protein assay with BSA (Sigma, UK) as standard. Ten micrograms of protein was loaded onto 12% Tris–glycine gels (Invitrogen, UK) and run under denaturing conditions at 100 V for 2 h. Electrophoresed proteins were transferred to Hybond-P nylon membrane (GE Healthcare, UK). Membranes were blocked with 5% skimmed milk (Marvel), then treated with primary DJ-1 monoclonal antibody (1:5000; Bioquote, UK), followed by HRP-conjugated secondary antibody. After a series of washes, the membranes were treated with ECL reagent (Pierce, UK) and autoradiographed on Kodak Biomax light films. A duplicate blot with identical protein

load was probed with beta-actin antibody (1:3000 dilution; Sigma, UK). Densitometric quantification of immunoblots was carried out using the Bio-Rad density measurement software. Results were expressed as the mean ratio of DJ-1/β-actin from 3 cases each. Analysis of sarkosyl-insoluble DJ-1 Sarkosyl insoluble proteins were isolated from brain tissue following the method of Goedert et al. (1992). In brief, frozen samples (1 g) of frontal cortex from the control, AD and FTLD cases were homogenised in buffer as described above. The homogenates were first centrifuged at 20,000×g and the super-

R. Kumaran et al. / Neurobiology of Disease 28 (2007) 122–132

125

Table 2 Immunohistochemical profiles of AT8, DJ-1, RD3 and RD4 in neurological disorders Disease/case no.

AD

CBD

PSP

FTLD with PBs

FTLD−MAPT mut R406W N279K Exon 10 + 16

Brain region

AT8-IR

DJ-1IR neurolesions PBs

NFTs

NTs

GIs

RD3-IR

RD4-IR

8 9 10 14 15 16 17 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

FC FC FC FC FC FC FC Po, FC, Hi Po Po, FC Po Po Po Po, FC, Hi Po, FC Po Po FC/TC/Hi FC FC FC TC FC FC FC

++++ ++++ +++ +++ +++ +++ +++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++

− − − − − − − − − − − − − − − − − +/++/++ ++ + ++ ++ ++ ++ +

++ ++ ++ − − − − 0/+ 0/+ 0/+ 0/+ 0/+ 0/+ 0/+ 0/+ 0/+ 0/+ − − − − − − − −

+++ ++ +++ − − − − 0/+ − 0/+ 0/+ 0/+ − 0/+ 0/+ 0/+ 0/+ − − − − − − − −

0/+ rare − 0/+ rare 0/+ (APs) 0/+ (APs) 0/+ (APs) + 0/+ (TA, CB) 0/+ (CB) 0/+ (TA, CB) 0/+ (CB) 0/+ (CB) 0/+ (CB) 0/+ (CB) 0/+ (TA, CB) 0/+ (CB) 0/+ (CB) − − − − − − − −

+ve +ve +ve − ve − ve − ve − ve − ve − ve − ve − ve − ve − ve − ve − ve − ve − ve +ve +ve +ve

+ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve +ve − ve − ve − ve

+ve +ve +ve +ve

− ve − ve − ve − ve

40 41 42 43 44

FC FC FC FC FC

+++ +++ ++++ ++++ ++++

− − − − −

++ ++ − − −

++ ++ 0/+ 0/+ 0/+

+ + ++ ++ ++

+ve +ve − ve − ve − ve

+ve +ve +ve +ve +ve

Semiquantitative estimations of AT8 IR and DJ-1 IR neurolesions in neurological disorders. RD3 IR and RD4 IR are mentioned as either present (+ve) or absent (− ve). Immunoreactivity was assessed semiquantitatively as − = absent, 0/+ = rare, + = few, ++ = moderate, +++ = many, ++++ = numerous. AP = astrocytic plaques, CB = coiled bodies, TA = tufted astrocyte, GI = glial inclusions, FC = frontal cortex, TC = temporal cortex, Po = Pons, Hi = hippocampus.

natant S1 was retained. The pellet P1 was rehomogenised in 5 vol of buffer and re-centrifuged. The two supernatants (S1 and S2) were combined and incubated with N-laurylsarkosine (1%) for 1 h at room temperature with shaking. Samples were then re-centrifuged for 1 h at 100,000×g. The resulting sarkosyl-insoluble pellets were resuspended (0.2 ml/g of starting material in 50 mM Tris– HCl, pH 7.4). Proteins were heated at 95 °C for 5 min prior to gel loading. Equal volumes of sarkosyl extracted fractions were run on 12% Tris–glycine gels (Invitrogen, UK) transferred onto hybond-P membranes (Amersham Biosciences) and probed with DJ-1 (BioQuote, UK) and AT8 (Innogenetics, UK) antibodies used at 1:1000 and 1:500 dilutions, respectively. Density of immunoreactive bands was measured as mentioned before. 2-dimensional gel electrophoresis (2DGE) Soluble protein homogenates (20 μg) from 2 control and two AD cases were applied to IPG strips (7 cm pI 4–7, Amersham Biosciences) and separated on IPGPhor (Amersham Biosciences) system according to manufacturer’s instructions. Second dimensional separation was performed by electrophoresis on 12% Tris–

glycine gels (Invitrogen, UK) and transferred onto hybond-P and probed with DJ-1 primary antibody as mentioned in the western blotting section. Results DJ-1 and tau; immunohistochemistry and double fluorescence confocal microscopy DJ-1 IR inclusions were observed in all AD cases, FTLD cases with Pick bodies and MAPT mutations, to a lesser extent in CBD and PSP cases, but were not seen in FTLD-U cases. Semiquantitative assessments of DJ-1 IR structures (compared to AT8positive structures) and 3R and 4R tau pathological changes in the range of neurological diseases studied here are summarised in Table 2. Sporadic tauopathies: AD, FTLD with Pick bodies, CBD and PSP In frontal cortex of all the AD cases (#8–10), DJ-1 IR was seen in extracellular plaques (Fig. 1b), in moderate numbers of

126

R. Kumaran et al. / Neurobiology of Disease 28 (2007) 122–132

Fig. 1. DJ-1 IR in human tauopathies. (a) AT-8 IR present in an extracelluar plaque in AD; (b) The same case showing DJ-1 IR NFTs and NTs associated with an extracellular plaque; (c) Another AD case showing DJ-1 IR in NFTs; (d, e) There is variable immunolabelling of Pick bodies with DJ-1 in two FTLD cases; (f) FTLD-case d showing RD3 positive Pick bodies; (g) DJ-1 IR glial cell inclusions in the pons in a PSP case; inset: DJ-1 IR tufted astrocyte in the frontal cortex from the same case; (h) Another PSP case showing variably labelled DJ-1 immunoreactive NFTs; (i) DJ-1 IR NTs associated with an astrocytic plaque in CBD. Panels a, b, c, e, g and i are from frontal cortex; d and f are from temporal cortex; h is from pontine base region. Scale bars: 10 μm.

intracellular NFTs (Fig. 1c) and more commonly in NTs (Table 2). In extracellular plaques, DJ-1 IR co-localised with plaque associated abnormal neurites and NFTs which were also AT8 IR (Figs. 1a, b). DFCM confirmed the co-localisation of the two proteins (Figs. 3a, b). All three cases had both 3R and 4R tau inclusions. In all three cases, numerous NFTs were stained by 4R tau whilst some NFTs and numerous NTs were positive for 3R tau. In all the FTLD with PBs cases examined, a proportion of PBs were positively labelled with DJ-1 IR in frontal and temporal cortical regions (Figs. 1d, e). Case 32 also demonstrated DJ-1 IR PBs in hippocampus. PBs positive for DJ-1 IR ranged from few to moderate when compared to AT8 IR (Table 2). All cases had Pick bodies that were typically positive for AT8, and these were also positive for 3R tau (Fig. 1f) but not 4R tau (see also de Silva et al., 2006). Co-localisation of DJ-1 and hyperphosphorylated tau was confirmed using DFCM (Fig. 3e). In one case, 3R tau immunoreactive pre-tangles were also present, though these were not DJ-1 IR. Overall, in PSP, only a few of AT8 tau positive structures showed DJ-1 IR (Table 2). These included NFTs and NTs, coiled bodies in the pontine base and frontal cortex (Figs. 1g, h). Some tufted astrocytes were also DJ-1 IR in frontal cortex (Fig. 1g); these were co-localised with tau (Figs. 3c, d). In CBD, some NTs associated with astrocytic plaques were DJ-1 IR (Fig. 1i) in frontal cortex in all 4 cases studied.

FTLD with MAPT mutations R406W mutation. Moderate numbers of NFTs and NTs in frontal cortex in the 2 FTLD cases with MAPT R406W mutation showed DJ-1 IR (Fig. 2a and Table 2). AT8-positive NFTs and NTs were positive for both 4R and 3R tau (Fig. 2b) (see also de Silva et al., 2006). Again, DFCM demonstrated co-localisation of DJ-1 and tau in NFTs and NTs. The DJ-1 inclusions appeared to form a perinuclear rim within affected neurons (Fig. 3f). N279K mutation. In the FTLD case with MAPT N279K mutation, DJ-1 IR was present in moderate number of glial cell inclusions within the cerebral cortical white matter (Fig. 2c) that resembled coiled bodies. These coiled bodies were 3R tau negative, but were positive for 4R tau isoform (Fig. 2d, see also de Silva et al., 2006). Additionally, some NTs were also DJ-1 IR, though NFT-like tau deposits within neurones were negative for DJ-1. Exon 10 + 16 mutation. The 2 FTLD cases with MAPT exon 10 + 16 mutation displayed a very similar staining pattern to that seen in FTLD with MAPT N279K mutation, with moderate numbers of AT8/4R tau positive glial cell inclusions in the cerebral cortical white matter showing DJ-1 IR (Figs. 2e, f and Table 2). FTLD-U. DJ-1 IR was not present in any ubiquitin inclusions in FTLD-U cases.

R. Kumaran et al. / Neurobiology of Disease 28 (2007) 122–132

127

Fig. 2. (a) DJ-1 IR in NFT and several NTs in FTLD case with MAPT R406W mutation. (b) The same case shows 3R positive tau inclusions in NFTs; (c) DJ-1 IR in coiled bodies in the FTLD MAPT N279K mutation case; inset: high magnification of a DJ-1 IR coiled body from the same case; (d) 4R tau immunoreactive coiled bodies in the FTLD MAPT N279 case; (e) DJ-1 IR coiled bodies in FTLD MAPT exon10 + 16 mutation case; inset: high magnification of a DJ-1 IR coiled body from the same case; (f) The same case demonstrating 4R tau immunoreactive coiled bodies. All panels are from frontal cortex region. Scale bars: in f, 10 μm and in insets, 5 μm.

Biochemical analysis

Discussion

2DGE. 2DGE analysis revealed multiple pI isoforms of DJ-1 ranging between 6.6 and 5.5. Some extra DJ-1 isoforms of pI 5.4 and 5.3 were present in small amounts in one control brain but was elevated in the two AD brains (Fig. 4a).

DJ-1 protein is expressed in significant quantities in human brain, suggesting an important physiological role. Our own previous immunohistochemical studies have shown that DJ-1 protein is expressed primarily in astrocytes in both control and PD cases, with only occasional neurons showing immunopositivity (Bandopadhyay et al., 2004). DJ-1 is also expressed by reactive astrocytes, and may be upregulated in inflammatory conditions and stroke (Lev et al., 2006 and our unpublished observations). Nevertheless, DJ-1 protein expression can be demonstrated in rodent neurons (Bandopadhyay et al., 2005; Bader et al., 2005; Kotaria et al., 2005), and DJ-1 mRNA expression can be demonstrated in human neurons and astrocytes by in-situ hybridisation (unpublished observations). Using new DJ-1 polyclonal antibodies, Olzmann et al. (2007) demonstrated DJ-1 IR in neurons in striatal and dopaminergic neurons in human brain. Similar observations were found in the primate brain. In the human brain, therefore, DJ-1 protein is likely to be expressed in neurons and glia and could have functional roles in both cell types. The presence of DJ-1 protein in pathological inclusions in human brain disease is further supporting evidence for neuronal expression of this protein. Although previous studies have shown that DJ-1 only rarely labels Lewy bodies, the pathological sig-

Analysis of soluble and insoluble DJ-1 in brain extracts. Western blots of homogenates from frozen frontal cortex showed that soluble DJ-1 protein levels were not significantly altered in AD or FTLD with Pick bodies when compared with control samples (Figs. 4b, c). To address whether fractions enriched in tau paired helical filaments in AD, or straight tau filaments in Pick bodies, contain DJ-1, sarkosyl-insoluble fractions were run on 10% Tris–glycine gels from two AD cases, two FTLD cases and 3 control cases. In these fractions, DJ-1 protein was enriched in AD and in FTLD cases, although a small amount of sarkosyl insoluble DJ-1 was also seen in three control cases. Immunoblots of the insoluble fraction probed with tau AT8 antibody demonstrated the characteristic pattern of tau bands in AD and FTLD brains, which were absent in the 3 control brains (Fig. 4d). Small amounts of high molecular weight DJ-1 species were also seen in the AD and FTLD brains, but were not present in controls (Fig. 4d).

128

R. Kumaran et al. / Neurobiology of Disease 28 (2007) 122–132

Fig. 3. Double confocal-immunofluorescence of DJ-1 and tau (AT8) proteins in AD, PSP, FTLD-Picks and FTLD with MAPT R406W cases. DJ-1 immunofluorescence is red, AT8 is green; yellow depicts a merged co-localisation of the two signals. (a) DJ-1 and tau are co-localised in NFTs and NTs in an extracellular plaque in an AD case; (b) The same AD case showing one NFT labelled with both DJ-1 and tau; (c) Co-localisation of DJ-1 and tau in NFT and (d) a coiled body in PSP; (e) Co-localisation of DJ-1 and tau proteins in two Pick bodies in FTLD; (f) Ring-like perinuclear deposition of DJ-1 and tau proteins in a MAPT R406W case in frontal cortex. Panels a, b, e and f are from frontal cortex; c and d are from pontine base.

nature of PD (Bandopadhyay et al., 2004; Meulener et al., 2005; Rizzu et al., 2004), DJ-1 has been identified in a subset of pathological tau inclusions in AD and FTLD (Meulener et al., 2005; Rizzu et al., 2004). The results of the present study confirm these latter findings, and demonstrate that DJ-1 protein is involved in the pathogenesis of a range of neurodegenerative diseases, mainly tauopathies. In the present study, DJ-1 was shown to be involved in the pathological changes of AD, FTLD with Pick bodies, and FTLD

with MAPT mutations, but most prominently in AD. DJ-1 IR was demonstrated in a subset of pathological inclusions in these conditions, occurring in moderate numbers of NFTs and NTs, and was also associated with extracellular amyloid plaques, and Pick bodies. DFCM confirmed the co-localisation of hyperphosphorylated tau and DJ-1 in these structures. An incompatibility of antigen retrieval techniques prevented the direct discrimination of 3R and 4R tau co-localisation with DJ-1 in this present study; however, localisation of both 3R and 4R tau in inclusions in AD

R. Kumaran et al. / Neurobiology of Disease 28 (2007) 122–132

129

Fig 4. (a) 2DGE pattern of DJ-1 protein pI isoforms from soluble fractions of 2 control and 2 AD cases. Please note extra acidic pI isoforms in AD brain (⁎); (b) Soluble DJ-1 protein levels in frontal cortex of FTLD-Picks, AD and control brains; (c) Semi-quantitative analysis of DJ-1 density levels normalised to beta-actin levels showed no difference between the three groups; (d) Sarkosyl-insoluble fractions from frontal cortex in FTLD with PBs (lanes 1, 2), AD (lanes 3, 4) and control (lanes 5–7) cases were assessed for insoluble DJ-1 and hyperphosphorylated tau. Phosphorylated tau was recovered in AD and FTLD with PBs, but not in controls. Small amounts of insoluble DJ-1 were recovered in control cases, but were enriched in insoluble tau preparations from AD and FTLD with PBs. Note the presence of small amounts of high molecular weight (47 kDa, 64 kDa) DJ-1 protein recovered in AD and FTLD-Picks brains but not in controls.

and other neuropathological conditions has been demonstrated (de Silva et al., 2003, 2006). The presence of both 3R and 4R tau isoforms in inclusion bodies in AD and other tauopathies is confirmed in the present study. In FTLD, a substantial number of Pick bodies showed variable labelling by DJ-1 using both immunohistochemistry with diaminobenzidine and double immunofluorescent labelling; these inclusions have been shown to contain specifically 3R tau (de Silva et al., 2003, 2006). Taken together, these findings support the view that DJ-1 can co-localise with both 3R and 4R tau isoforms, and provide confirmation of two previous reports (Neumann et al., 2004; Rizzu et al., 2004) showing that DJ-1 positive inclusions are co-localised with tau in AD and FTLD. Analysis of DJ-1 IR in cases of FTLD with various MAPT mutations revealed interesting patterns of DJ-1 IR (see Table 2). In R406W MAPT mutation, DJ-1 was present in a substantial number of NFTs and NT’s. These R406W cases have both 3R and 4R tau isoform containing pathological inclusions (de Silva et al., 2006). However, in the N279K MAPT mutation case, DJ-1 IR was present in a number of glial cell inclusions in the cerebral cortical white matter that resembled coiled bodies, but was not seen in NFTs. In this case both neuronal and glial inclusions have been shown to be

positive for 4R tau (de Silva et al., 2006 and personal observations). Similarly, DJ-1 was present in moderate numbers of glial cell inclusions, these again resembling coiled bodies, in the two cases with MAPT exon 10 + 16 splice site mutation. These cases are 3R tau negative, and therefore, in these cases too DJ-1 is present in 4R tau positive glial cell inclusions. The reason for DJ-1 associating with both 3R and 4R tau isoforms in subsets of glial and neuronal structures is unclear. However, we do not think that these results are due to DJ-1 antibody sticking non-specifically to tau molecules as the same antibody did not recognise recombinant tau in immunoblots (data not shown). Our present data are speculative of a chaperone role for DJ-1 with both 3R and 4R tau. The structure of DJ-1 resembles the molecular chaperone Hsp31 (Lee et al., 2003). Therefore, one of the functions of DJ-1 could be to offset the misfolding of tau (Neumann et al., 2004). However, if this is so, it is unclear as to why DJ-1 is not present in all tau positive structures in AD, FTLD or PSP and CBD. Nevertheless, we speculate that DJ-1 associates with tau when it is undergoing beta-sheet formation in order to unwind it and then as the disease progresses, DJ-1 is sequestered in some NFTs and NTs. More direct evidence perhaps from in vitro

130

R. Kumaran et al. / Neurobiology of Disease 28 (2007) 122–132

experiments is necessary to prove this, using tau as a substrate. Interestingly, two reports that investigated the chaperone role of DJ-1 in vitro using alpha-synuclein as a substrate, arrived at slightly different conclusions regarding the cysteine residues crucial for activation of oxidation driven chaperone activity (Shendelman et al., 2004; Zhou et al., 2006). In addition, Zhou et al. (2006) suggested that overoxidation of DJ-1 protein might lead to loss of its ability to inhibit alpha-synuclein fibrillation. In this respect, our data on additional oxidised isoforms of DJ-1 seen in AD brains (Fig. 4a) might reflect a loss of the putative chaperone function against tau. The absence of DJ-1 in majority of Lewy bodies, however, argues against the direct involvement of DJ-1 and alpha-synuclein in human brain neurons although the two proteins co-localised in a subpopulation of oligodendroglial inclusions in MSA (Neumann et al., 2004). However, DJ-1 IR was not associated with ubiquitin inclusions in FTLD-U inclusions, demonstrating a specificity of association of DJ-1 with tau inclusions in these diseases. Other possible reasons for DJ-1 associating with tau inclusions could include oxidative stress or age-related modifications in the structure and properties of DJ-1 (Bandopadhyay et al., 2004; Meulener et al., 2006). A recent study has shown that DJ-1 can downregulate phosphatase and tensin homologue deleted on chromosome 10 (PTEN)—a tumor suppressor (Kim et al., 2005b). PTEN exerts its activity by inhibiting PKB/Akt mediated intracellular signalling pathway and tau can be phosphorylated by Akt (Ksiezak-Reding et al., 2003). Related to this, two recent studies have linked PTEN/Akt pathway with AD pathology (Griffin et al., 2005; Pei et al., 2003). Additionally, inactivation of DJ-1 in Drosophila leads to impaired PI3K/Akt signalling (Yang et al., 2005). Therefore DJ-1 could directly or indirectly affect PTEN/Akt/tau phosphorylation pathways. These modifications could manifest in the appearance of DJ-1 in NFTs, NTs and glial inclusions in vulnerable cell populations in AD, FTLD and tauopathies. Interestingly, DJ-1 is not the only familial PD causing protein that has been found to associate with hyperphosporylated tau; a recent study has demonstrated LRRK2 (PARK8) in NFTs and NTs in AD and other tauopathies (Miklossy et al., 2006). Differential fractionation and western blotting techniques demonstrated that the levels of soluble DJ-1 monomer in AD and FTLD were similar to those in controls, but in AD and FTLD, the sarkosyl-insoluble fraction was shown to be enriched in DJ-1. These conclusions are provisional because the data are derived from a small number of cases (n = 3) and need to be replicated in a larger case series. The AD cases, nonetheless, displayed typical patterns of tau bands, showing that the extraction procedure was valid. Some high molecular weight DJ-1 was recovered in the sarkosyl insoluble fractions in the AD and FTLD cases and may reflect more insoluble DJ-1 in these cases. The nature of high molecular weight DJ-1 species is unknown but could include oxidative modifications to DJ-1 protein (Bandopadhyay et al., 2004; Choi et al., 2006). Moreover, small amounts of SDS resistant high molecular weight DJ-1 species have been shown to accumulate in MSA cerebellum (Neumann et al., 2004). In the present study, the amount of insoluble DJ-1 monomer was approximately 3 and 4 times that of control levels in FTLD-Pick’s and AD respectively (Fig. 4d). It is possible that the ageing process itself could make DJ-1 insoluble, and this might explain the presence of DJ-1, albeit, in small amounts, in insoluble fractions in control subjects. Our 2DGE data demonstrate that like our previous observations in PD brains (Bandopadhyay et al., 2004), DJ-1 in AD brains is also subject to oxidative stress damage (Fig. 4) compared to control

brains; these results are in accordance with Choi et al. (2006). It is possible that the oxidative damage to DJ-1 protein may inactivate its function. Conclusion Present findings indicate a differential expression of DJ-1 in pathological inclusions in various neurodegenerative disorders and support a diverse role for DJ-1 in the pathogenesis of a variety of tauopathies. The data suggest that DJ-1 could play a role in the pathogenesis of AD, FTLD-Pick’s and FTLD-MAPT mutations, a much lesser role in PSP, CBD and none in FTLD-U. In some instances, DJ-1 was present in a variety of neuronal inclusions, whilst in others it was present in glial cell inclusions. DJ-1 was present in only a subset of hyperphosphorylated, tau-positive pathological structures and was associated with both 3R and 4R tau isoform containing inclusions. In addition we demonstrated that DJ-1 protein undergoes additional changes in its pI isoforms in AD brain and this may reflect upon its normal function. Unravelling the molecular role of DJ-1 in the brain will increase our understanding of neurodegenerative diseases, and may point towards potential neuroprotective strategies. Acknowledgments The authors RB, AK and RK thank Parkinson’s disease Society of UK (PDS UK), the PDS UK SPRING and the Reta Lila Weston Institute for funding this work and the Queen Square Brain Bank for tissue. The Manchester Brain Bank is supported in part by the Alzheimer’s Research Trust. References Abou-Sleiman, P.M., Healy, D.G., Quinn, N., Lees, A.J., Wood, N.W., 2003. The role of pathogenic DJ-1 mutations in Parkinson’s disease. Ann. Neurol. 54, 283–286. Annesi, G., Savettieri, G., Pugliese, P., D’Amelio, M., Tarantino, P., Ragonese, P., La Bella, V., Piccoli, T., Civitelli, D., Annesi, F., Fierro, B., Piccoli, F., Arabia, G., Caracciolo, M., Ciro Candiano, I.C., 2005. DJ-1 mutations and parkinsonism–dementia–amyotrophic lateral sclerosis complex. Ann. Neurol. 58, 803–807. Bader, V., RAn ZHu, X., Lubbert, H., Stichel, C.C., 2005. Expression of DJ-1 in the adult mouse CNS. Brain Res. 1041, 102–111. Bandopadhyay, R., Kingsbury, A.E., Cookson, M.R., Reid, A.R., Evans, I.M., Hope, A.D., Pittamn, A., Lashley, T., Canet-Aviles, R., Miller, D.W., McLendon, C., Strand, C., Leonard, A.J., Abou-Sleiman, P.M., Healy, D.G., Ariga, H., Wood, N.W., de Silva, R., Revesz, T., Hardy, J.A., Lees, A.J., 2004. The expression of DJ-1 (PARK7) in normal human CNS and idiopathic Parkinson’s disease. Brain 127, 420–430. Bandopadhyay, R., Miller, D.W., Kingsbury, A.E., Jowett, T.P., Kaleem, M.M., Pittman, A.M., de Silva, R., Cookson, M.R., Lees, A.J., 2005. Development, characterisation and epitope mapping of novel monoclonal antibodies for DJ-1 (PARK7) protein. Neurosci. Lett. 383, 225–230. Bandyopadhyay, S., Cookson, M.R., 2004. Evolutionary and functional relationships within the DJ-1 family. BMC Biol. 4, 6. Bonifati, V., Rizzu, P., van Baren, M.J., Schaap, O., Breedveld, G.J., Krieger, E., Dekker, M.C., Squitieri, F., Ibanez, P., Joosse, M., van Dongen, J.W., Vanacore, N., van Swieten, J.C., Brice, A., Meco, G., van Dujin, C.M., Oostra, B.A., Heutink, P., 2003. Mutations in the DJ-1 gene associated with autosomal recessive early-onset parkinsonism. Science 299, 256–259. Chen, L., Cagniard, B., Mathews, T., Jones, S., Koh, H.C., Ding, Y., Carvey,

R. Kumaran et al. / Neurobiology of Disease 28 (2007) 122–132 P.M., Ling, Z., Kang, U.J., Zhuang, X., 2005. Age-dependent motor deficits and dopaminergic dysfunction in DJ-1 null mice. J. Biol. Chem. 280, 12426–21418. Choi, J., Cameron Sullards, M., Olzmann, J.A., Rees, H.D., Weintraub, S.T., Bostwick, D.E., Gearing, M., Levey, A.I., Chin, L.S., Li, L., 2006. Oxidative damage of DJ-1 is linked to sporadic Parkinson’s and Alzheimer diseases. J. Biol. Sci. 281, 10816–10824. de Silva, R., Lashley, T., Gibb, G., Hanger, D., Hope, A., Reid, A., Bandopadhyay, R., Utton, M., Strand, C., Jowett, T., Khan, N., Anderton, B., Wood, N., Holton, J., Revesz, T., Lees, A., 2003. Pathological inclusion bodies in tauopathies contain distinct complements of tau with three or four microtubule-binding repeat domains as demonstrated by new specific monoclonal antibodies. Neuropathol. Appl. Neurobiol. 29, 288–302. de Silva, R., Lashley, T., Strand, C., Shiarli, A.M., Shi, J., Tian, J., Bailey, K.L., Davies, P., Bigio, E.H., Arima, K., Iseki, E., Murayama, S., Kretzschmar, H., Neumann, M., Lippa, C., Halliday, G., MacKenzie, J., Ravid, R., Dickson, D., Wszolek, Z., Iwatsubo, T., Pickering-Brown, S.M., Holton, J., Lees, A., Revesz, T., Mann, D.M.A., 2006. An immunohistochemical study of cases of sporadic and inherited frontotemporal lobar degeneration using 3R- and 4R-specific tau monoclonal antibodies. Acta Neuropathol. (Berl) 111, 329–340. Galpern, W.R., Lang, A.E., 2006. Interface between tauopathies and synucleinopathies: a tale of two proteins. Ann. Neurol. 59 (3), 449–458. Goedert, M., 2005. Tau gene mutations and their effects. Mov. Disord. 20 (Suppl. 12), S45–S52. Goedert, M., Spillantini, M.G., Cairns, N.J., Crowther, R.A., 1992. Tau proteins of Alzheimer paired helical filaments: abnormal phosphorylation of all six brain isoforms. Neuron 8, 159–168. Goldberg, M.S., Pisani, A., Haburcak, M., Vortherms, T.A., Kitada, T., Costa, C., Tong, Y., Martella, G., Tscherter, A., Martins, A., Bernardi, G., Roth, B.L., Pothos, E.N., Calabresi, P., Shen, J., 2005. Nigrostriatal dopaminergic deficits and hypokinesia caused by inactivation of the familial Parkinsonism-linked gene DJ-1. Neuron 45, 489–496. Griffin, R.J., Moloney, A., Kelliher, M., Johnson, J.A., Ravid, R., Dockery, P., O’Connor, R., O’Neill, C., 2005. Activation of Akt/PKB, increased phosphorylation of Akt substrates and loss and altered distribution of Akt and PTEN are features of Alzheimer’s disease pathology. J. Neurochem. 93, 105–117. Hague, S., Rogaeva, E., Hernandez, D., Gullick, C., Singleton, A., Hanson, M., Johnson, J., Weiser, R., Gallardo, M., Ravina, B., Gwinn-Hardy, K., Crawley, A., St George-Hyslop, P.H., Lang, A.E., Heutink, P., Bonifati, V., Hardy, J., 2003. Early-onset Parkinson’s disease caused by a compound heterozygous DJ-1 mutation. Ann. Neurol. 54, 271–274. Hedrich, K., Djarmati, A., Schafer, N., Hering, R., Wellenbrock, C., Weiss, P.H., Hilker, R., Vieregge, P., Ozelius, L.J., Heutink, P., Bonifati, V., Schwinger, E., Lang, A.E., Noth, J., Bressman, S.B., Pramstaller, P.P., Reiss, O., Klein, C., 2004. DJ-1 (PARK7) mutations are less frequent than Parkin (PARK2) mutations in early-onset Parkinson disease. Neurology 62, 389–394. Hod, Y., Pentyala, S.N., Whyard, T.C., El-Maghrabi, M.R., 1999. Identification and characterization of a novel protein that regulates RNA–protein interaction. J. Cell. Biochem. 72, 435–444. Huai, Q., Sun, Y., Wang, H., Chin, L.S., Li, L., Robinson, H., Ke, H., 2003. Crystal structure of DJ-1/RS and implication on familial Parkinson’s disease. FEBS Lett. 549, 171–175. Kim, R.H., Smith, P.D., Aleyasin, H., Hayley, S., Mount, M.P., Pownall, S., Wakeham, A., You-Ten, A.J., Kalia, S.K., Horne, P., Westaway, D., Lozano, A.M., Anisman, H., Park, D.S., Mak, T.W., 2005a. Hypersensitivity of DJ-1-deficient mice to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyrindine (MPTP) and oxidative stress. Proc. Natl. Acad. Sci. USA. 102, 5215–5220. Kim, R.H., Peters, M., Jang, Y.J., Shi, W., Pintillie, M., Fletcher, G.C., DeLuca, C., Liepa, J., Zhou, L., Snow, B., Binari, R.C., Manoukian, A.S., Bray, M.R., Liu, F.F., Tsao, M., Mak, T.W., 2005b. DJ-1, a novel regulator of the tumor suppressor PTEN. Cancer Cell 7, 263–273. Kotaria, N., Hinz, U., Zechel, S., von Bohlen, U.N.D., Halbach, O., 2005.

131

Localisation of DJ-1 protein in the murine brain. Cell Tissue Res. 322, 503–507. Ksiezak-Reding, H., Pyo, H.K., Feinstein, B., Pasinetti, G.M., 2003. Akt/ PKB kinase phosphorylates separately Thr212 and Ser214 of tau protein in vitro. Biochim. Biophys. Acta 1639, 159–168. Lee, S.J., Kim, S.J., Kim, I.K., Ko, J., Jeong, C.S., Kim, G.H., Park, C., Kang, S.O., Suh, P.G., Lee, H.S., Cha, S.S., 2003. Crystal structures of human DJ-1 and Escherichia coli Hsp31, which share an evolutionarily conserved domain. J. Biol. Chem. 278, 44552–44559. Lev, N., Ickowicz, D., Barhum, Y., Blondheim, N., Melamed, E., Offen, D., 2006. Experimental encephalomyelitis induces changes in DJ-1: implications for oxidative stress in multiple sclerosis. Antioxid. Redox Signal. 8, 1987–1995. Martinat, C., Shendelman, S., Jonason, A., Leete, T., Beal, M.F., Yang, L., Floss, T., Abeliovich, A., 2004. sensitivity to oxidative stress in Dj-1 deficient dopamine neurons: an ES-derived cell model of primary parkinsonism. Plos Biol. 2 (11), 327. Meulener, M.C., Graves, C.L., Sampatu, D.M., Armstrong-Gold, C.E., Bonini, N.M., Giasson, B.I., 2005. DJ-1 is present in a large molecular complex in human brain tissue and interacts with alpha-synuclein. J. Neurochem. 93, 1524–1532. Meulener, M.C., Xu, K., Thomson, L., Ischiropoulos, H., Bonini, N.M., 2006. Mutational analysis of DJ- in Drosophila implicates functional inactivation by oxidative damage and aging. Proc. Natl. Acad. Sci. 103, 12517–12522. Miklossy, J., Arai, T., Guo, J.P., Klegeris, A., Yu, S., McGeer, E.G., McGeer, P.L., 2006. LRRK2 expression in normal and pathologic human brain in human cell lines. J. Neuropathol. Exp. Neurol. 12, 521–528. Miller, D.W., Ahmad, R., Hague, S., Baptista, M.J., Canet-Aviles, R., McLendon, C., Carter, D.M., Zhu, P.P., Stadler, J., Chandran, J., Klinefelter, G.R., Blackstone, C., Cookson, M.R., 2003. L166P mutant DJ-1, causative for recessive Parkinson’s disease, is degraded through the ubiquitin–proteasome system. J. Biol. Chem. 278, 36588–36595. Mitsumoto, A., Nakagawa, Y., 2001. DJ-1 is an indicator for endogenous reactive oxygen species elicited by endotoxin. Free Radic. Res. 35, 885–893. Nagakubo, D., Taira, T., Kitaura, H., Ikeda, M., Tamai, K., Iguchi-Ariga, S.M., Ariga, H., 1997. DJ-1, a novel oncogene which transforms mouse NIH3T3 cells in cooperation with ras. Biochem. Biophys. Res. Commun. 231, 509–513. Neumann, M., Muller, V., Gorner, K., Kretzschmar, H.A., Haas, C., Kahle, P.J., 2004. Pathological properties of the Parkinson’s disease-associated protein DJ-1 in alpha-synucleinopathies and tauopathies: relevance for multiple system atrophy and Pick’s disease. Acta Neuropathol. (Berl) 107, 489–496. Olzmann, J.A., Bordelon, J.R., Muly, E.C., Rees, H.D., Levey, A.I., Li, L., Chin, L.S., 2007. Selective enrichment of DJ-1 protein in primate striatal neuronal processes: implications for parkinson’s disease. J. Comp. Neurol. 500, 585–599. Pei, J.J., Khatoon, S., An, W.L., Nordlinder, M., Tanaka, T., Braak, H., Tsujio, I., Takeda, M., Alafuzoff, I., Winblad, B., Cowburn, R.F., Grundke-Iqbal, I., Iqbal, K., 2003. Role of protein kinase B in Alzheimer’s neurofibrillary pathology. Acta. Neuropathol. (Berl) 105, 381–392. Pisani, A., Martella, G., Tscherter, A., Costa, C., Mercuri, N.B., Bernardi, G., Shen, J., Calbresi, P., 2006. Enhanced sensitivity of DJ-1 deficient dopaminergic neurons to energy metabolism impairment: role of Na+/K+ ATPase. Neurobiol. Dis. 23, 54–60. Rizzu, P., Hinkle, D.A., Zhukareva, V., Bonifati, V., Severijnen, L.A., Martinez, D., Ravid, R., Kamphorst, W., Eberwine, J.H., Lee, V.M., Trojanowski, J.Q., Heutinck, P., 2004. DJ-1 colocalizes with tau inclusions: a link between parkinsonism and dementia. Ann. Neurol. 55, 113–118. Shendelman, S., Jonason, A., Martinat, C., Leete, T., Abeliovich, A., 2004. DJ-1 is a redox-dependent molecular chaperone that inhibits alphasynuclein aggregate formation. Plos Biol. 2, 1764–1773. Taira, T., Saito, Y., Niki, T., Iguchi-Ariga, S.M., Takahashi, K., Ariga, H.,

132

R. Kumaran et al. / Neurobiology of Disease 28 (2007) 122–132

2004. DJ-1 has a role in antioxidative stress to prevent cell death. EMBO Rep. 5, 213–218. Tao, X., Tong, L., 2003. Crystal structure of human DJ-1, a protein associated with early onset Parkinson’s disease. J. Biol. Chem. 278, 31372–31379. Wagenfeld, A., Gromoll, J., Cooper, T.G., 1998. Molecular cloning and expression of rat contraception associated protein 1 (CAP1), a protein putatively involved in fertilization. Biochem. Biophys. Res. Commun. 251, 545–549. Wei, Y., Ringe, D., Wilson, M.A., Onderchen, M.J., 2007. Identification of functional subclasses in the DJ-1 superfamily proteins. PloS Comput. Biol. 3, e10.

Wilson, M.A., Collins, J.L., Hod, Y., Ringe, D., Petsko, G.A., 2003. The 1.1-A resolution crystal structure of DJ-1, the protein mutated in autosomal recessive early onset Parkinson’s disease. Proc. Natl. Acad. Sci. USA. 100, 9256–9261. Yang, Y., Gehrke, S., Haque, M.D.E., Imai, Y., Kosek, J., Yang, L., Flint Beal, M., Nishimura, I., Wakamatsu, K., Ito, S., Takahashi, R., Lu, B., 2005. Inactivation of Drosophila DJ-1 leads to impairments of oxidative stress response and phosphatidylinositol 3-kinase/Akt signalling. Proc. Natl. Acad. Sci. 102, 13670–13675. Zhou, W., Zhu, M., Wilson, M.A., Petsko, G.A., Fink, A.L., 2006. The oxidation state of DJ-1 regulates its chaperone activity towards alphasynuclein. J. Mol. Biol. 356 (4), 1036–1048.