- Email: [email protected]
Localization of DJ-1 mRNA in the mouse brain
Neuroscience Letters 367 (2004) 273–277
Localization of DJ-1 mRNA in the mouse brain Huifang Shang a,b , Doris Lang c , Burgunder Jean-Marc b,d , Alain Kaelin-Lang b,∗ a Department of Neurology, West China Hospital, SiChuan University, 610041 Chengdu, SiChuan, China Department of Neurology, Laboratory of Neuromorphology, Inselspittal, University of Bern, CH-3010 Bern, Switzerland c Neuro-Oncology Program, Division of Oncology, University Children’s Hospital Zurich, CH-8032 Zurich, Switzerland d Department of Medicine, National University of Singapore, 5 Lower Kent Ridge Road, Singapore 119074, Singapore
b
Received 13 February 2004; received in revised form 1 June 2004; accepted 2 June 2004
Abstract DJ-1 is mutated in autosomal recessive, early onset Parkinson’s disease but the exact localization of the DJ-1 gene product in the mammalian brain is largely unknown. We aimed to evaluate the DJ-1 mRNA expression pattern in the mouse brain. Serial coronal sections of brains of five male and five female adult mice were investigated by using in situ hybridization with a DJ-1 specific 35 S-labeled oligonucleotide probe. Hybridized sections were analyzed after exposure to autoradiography films and after coating with a photographic emulsion. DJ-1 was heterogeneously expressed throughout the mouse central nervous system. A high expression of DJ-1 mRNA was detected in neuronal and non-neuronal populations of several structures of the motor system such as the substantia nigra, the red nucleus, the caudate putamen, the globus pallidus, and the deep nuclei of the cerebellum. Furthermore, DJ-1 mRNA was also highly expressed in non-motor structures including the hippocampus, the olfactory bulb, the reticular nucleus of the thalamus, and the piriform cortex. The high expression of DJ-1 mRNA in brain regions involved in motor control is compatible with the occurrence of parkinsonian symptoms after DJ-1 mutations. However, expression in other regions indicates that a dysfunction of DJ-1 may contribute to additional clinical features in patients with a DJ-1 mutation. © 2004 Elsevier Ireland Ltd. All rights reserved. Theme: Motor systems and sensorimotor integration Topic: Basal ganglia, cortex, cerebellum Keywords: DJ-1 gene; Parkinson’s disease; In situ hybridization; Central nervous system
The DJ-1 gene was first identified as an oncogene [12] and contains eight exons spanning 24-kb. This gene encodes a highly conserved protein (DJ-1). Recently, mutations in the DJ-1 gene have been found to be associated with autosomal recessive, early onset Parkinson’s disease (PD) [3,8]. A 14-kb homozygous deletion in the DJ-1 gene, which removes exons 1 through 5, was identified to segregate with PD in a large Dutch family [3], and a homozygous 497T > C transition, which results in a missense mutation (L66P), was detected to segregate with PD in an Italian family [3]. Two other point mutations, R98Q and A104T, were identified in PD patients [8]. Two mutations (IVS6-1G → C and c.56delG c.57G → A), which were predicted to drastically change the protein product and result in the dysfunction of ∗ Corresponding author. Tel.: +41-31-632-21-11; fax: +41-31-632-27-52. E-mail address: [email protected] (A. Kaelin-Lang).
the protein, were also reported [8]. The mechanism by which altered DJ-1 induces PD is not known so far but several hypotheses have been presented [2]. Altogether, DJ-1 mutations account for a small number of autosomal recessive juvenile Parkinson cases [1]. In the human brain, northern blot studies revealed a ubiquitous expression of the DJ-1 gene particularly in subcortical structures, such as the caudate nucleus, the hippocampus, the thalamus, and the substantia nigra [3]. Using immunohistochemistry, abundant DJ-1 protein immunoreactivity was found in the frontal cortex and in the substantia nigra of normal, PD and PSP brain, and was mainly located in astrocytes [2]. However, the exact distribution and cellular localization of the DJ-1 mRNA, especially in other regions of the central nervous system is still unknown. The aim of the present study was to map the expression of the DJ-1 gene in the mouse brain by using in
0304-3940/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2004.06.002
274
H. Shang et al. / Neuroscience Letters 367 (2004) 273–277
situ hybridization (ISH) histochemistry with a 35 S-labeled oligonucleotide probe. Five male and five female adult mice were housed in a light- and temperature-controlled room (12-h dark:12-h light cycles) with normal mice chow and water ad libitum. These mice were decapitated under ether anesthesia. Their brains were then frozen in isopentane (at −40 ◦ C) and stored at −70 ◦ C. Serial coronal sections (12 m) of the entire brain were cut on a cryostat-microtome (Reichert-Jung, Germany) at −20 ◦ C and mounted onto gelatin-coated slides. The oligonucleotide probe for mouse DJ-1 mRNA was designed using the software Primer Designer 4 (Sci-ed software, Durham, USA; GenBank No.: AB015652). The sequence of the DJ-1 probe (anti-sense) is 5 ACAATGGCTAGTGCAAACTCAAAGCTGGTCCCCGGCCCGC-3 , which is complementary to the nucleotides 569–608 of mouse DJ-1 mRNA. As a control, a sense probe of the same nucleotides was used. The probes were commercially synthesized (Microsynth AG, Balgach, Switzerland). They were labeled with 35 S-deoxyadenosine-alpha-thiotriphosphate (35 S-␣-dATP, Hartmann Analytic, Braunschweig, Germany) using terminal deoxynucleotidyl transferase (Roche Diagnostics GmbH, Mannheim, Germany) according to the standard 3 end labeling reaction protocol of the manufacturer. The labeling reaction was stopped on ice and with 400 l TE buffer (0.1 M Tris–HCl pH 7.5, 10 mM EDTA), extracted and purified by phenol and chloroform, precipitated by ethanol and sodium chloride (NaCl) as described elsewhere [4]. Hybridization histochemistry was performed as previously described [4]. In brief, the sections were fixed with 4% formaldehyde in phosphate-buffered saline (PBS, pH 7.4) for 6 min. After washing twice with PBS, they were placed in 0.25% acetic anhydride in 0.1 M triethanolamine/0.9% NaCl (pH 8), for 10 min and dehydrated through a graded series of ethanol (70–100%). Sections were immersed in chloroform followed by a partial rehydration in ethanol (100–95%), and air dried at room temperature (RT). The slides were incubated with 35 S-labeled probes in 50 l of hybridization buffer containing 50% formamide (v/v), 600 mM NaCl, 80 mM Tris–HCl (pH 7.5), 4 mM EDTA, 0.1% (w/v) sodium pyrophosphate, 0.2% (w/v) SDS, 50 IU/ml Heparin, and 10% (w/v) Na-dextran sulfate at 37 ◦ C for 20 h. Posthybridization washing was performed in 0.5× SSC (1× SSC = 0.15 M NaCl/0.015 M sodium citrate) at 55 ◦ C for 4 × 15 min and in 0.5× SSC at RT for 2 × 30 min. The slides were air dried and exposed to Kodak BioMax MRI film (Eastman Kodak Company, Rochester, New York, USA) for 7 days. The autoradiography films were analyzed using the NIH Image software. The slides were then immersed in Ilford K5 nuclear emulsion (Ilford, Mobberly, Cheshire, UK) diluted in water (2:1, v/v) and exposed for 4 weeks at 4 ◦ C in darkness before being developed with Kodak D-19 for 4 min at 20 ◦ C, then fixed in Ilford Hypam fixer (Ilford; 1:4, v/v) for 4 min. After washing, the sections were counterstained with 0.2% toluidine
blue. The DJ-1 mRNA expression of the ISH experiments was evaluated according to silver grain densities on the sections. Signals were assessed as strongly positive, moderate, weakly positive, and not detectable and the localization was identified according to a standard mouse brain atlas [13]. Specificity of the probe and the hybridization procedure were checked as follows: (1) incubation with the 35 S-labeled sense probe; (2) pretreatment with ribronuclease (RNase) containing 20 g/ml RNase A (Sigma-Aldrich Corp., St. Louis, MO, USA) and 1 unit/ml RNase T1 (Sigma); (3) co-incubation with labeled and a 100-fold excess of unlabeled DJ-1 probe; and (4) co-incubation with a mixture of sense and anti-sense DJ-1 probe. Our results showed a similar DJ-1 mRNA expression pattern in all ten brains. The examination of the autoradiography films hybridized with the anti-sense DJ-1 oligonucleotide probe displayed a widespread expression (Fig. 1A–E). At the cellular level, examination of emulsion-coated tissue sections revealed a high number of specific silver grain clusters overlying toluidine blue counterstained cells with a typical neuronal morphology. No signal was seen on the film (Fig. 1G–I) and on emulsion-coated tissue sections processed with RNase pretreatment, sense DJ-1 probe, co-incubation with labeled and a 100-fold excess of unlabeled DJ-1 probe, and co-incubation with a mixture of sense and anti-sense DJ-1 probe. In both the olfactory bulb (Fig. 1F) and the lobules of cerebellum (Fig. 1J), a weak signal was seen with the sense probe. However this signal was much lower than when the antisense probe was used. Several structures involved in the regulation of motor activity had a high level of expression: neurons in the motor cortex (Fig. 2A), the caudate putamen (Fig. 2B), the substantia nigra (Fig. 2C), and the red nucleus (Fig. 2D) were strongly labeled. The cerebellum also exhibited strong DJ-1 expression (Fig. 1E), particularly in neurons of the granule cell layers and the deep cerebellar nuclei (Fig. 2E), and in Purkinje cells (data not shown). In contrast, the molecular cell layer in the cerebellum was only weakly labeled (data not shown). In addition, we also found several strongly labeled large cells in the lateral globus pallidus (data not shown) and other nuclei such as trigeminal motor neurons (Fig. 2F) in the brain stem. Other structures were also strongly labeled. In particular, strong hybridization signals were detected in the hippocampus (Fig. 1C and D), the piriform cortex (Fig. 1B), the olfactory cortex (Fig. 1A), and the reticular nucleus of the thalamus (Fig. 1C). By contrast, the cells in the secondary somatosensory cortex (data not shown) were only weakly labeled. In the brain structures with a high DJ-1 expression level and a high neuronal density, like the hippocampus, the olfactory bulb, and the cerebellum, the morphology of individual DJ-1 positive cells could not be assessed accurately. In all other structures, most of the DJ-1 positive cells had a typical neuronal morphology. A weaker more diffuse but specific signal was also observed between these neuronal cells
H. Shang et al. / Neuroscience Letters 367 (2004) 273–277
275
Fig. 1. Film autoradiogram showing the distribution of the DJ-1 mRNA in the adult mouse brain. Coronal sections hybridized with the radiolabeled oligonucleotide anti-sense probe (A–E) and sense probe (F–J) for DJ-1. Regions with high DJ-1 mRNA expression include olfactory bulb (A), caudate putamen (Cpu; B and C), piriform cortex (Pir; B), hippocampus (Hippo; C), lateral globus pallidus (LGP; C), medial globus pallidus (MGP; C), reticular thalamic nucleus (Rt; C), red nucleus (RN; D), substantia nigra (SN; D), amygdalohippocampal area (AHi; D), cerebellar cortex and dentate nucleus (DCN; E). Scale bar, 1 mm.
276
H. Shang et al. / Neuroscience Letters 367 (2004) 273–277
Fig. 2. Photomicrographs of emulsion-dipped sections depicting the cellular mRNA expression of DJ-1 in neurons of the mouse brain (bright field microscopy, counterstained with toluidine blue). (A) Motor cortex, (B) caudate putamen nucleus, (C) substantia nigra, (D) red nucleus, (E) dentate nucleus of the cerebellum, and (F) trigeminal motor nucleus. Scale bars, 100 m.
(Fig. 2). In contrast, no specific signal was detected on the surrounding white matter. In addition, other non-neuronal structures like the choroid plexus also exhibited DJ-1 expression. An important finding of our study is the demonstration of a widespread and strong DJ-1 expression in neuronal populations of several structures of the motor system. The high DJ-1 expression in neurons of the caudate putamen and substantia nigra of the mouse brain confirms previous results in human brain homogenates [3]. Since basal ganglia, thalamus, and motor cortex are structures involved in the pathophysiology of PD in humans, we can speculate that mutated DJ-1 could predominantly impair the function of these structures and thus lead to the development of PD symptoms. However, further studies are needed to investigate possible species difference in the pattern of DJ-1 gene expression. A recent study using immunohistochemistry on human brains demonstrated a stronger DJ-1 immunoreactivity in astrocytes than in neurons [2]. Also in our study we found a specific DJ-1 expression between neuronal cells that could be due to glial expression. However, further studies are needed to quantitatively investigate the level of expression as well as the subcellular localization of DJ-1 expression in neurons and glial cells. Although the mechanism by which mutated DJ-1 in neurons or astrocytes causes PD is unclear [2,3,10], recent studies show that DJ-1 may be involved in the cellular response to oxidative stress. The mutant DJ-1 protein may result in conformational changes of DJ-1, causing the mislocalization of DJ-1 to mitochondria or decrease the interaction of DJ-1 with the binding partners, and finally
decrease the ability to limit oxidative damage [3,10,11]. Since oxidative damage has been found to be involved in death of the nigral dopaminergic neurons and in the pathogenesis of PD [7], it is conceivable that the mutation of DJ-1 leads to the development of PD through oxidative stress. Furthermore, we found a strong DJ-1 expression in neurons of the cerebellum. This finding confirms and extends the previous report of moderate DJ-1 gene expression in tissue homogenates of the human cerebellum [3]. The clinical importance of this finding is unclear. Cerebellar symptoms have not been described in PD patients with DJ-1 mutation. However, a recent clinical study showed a possible cerebellar involvement in DJ-1 linked PD by assessing cerebral glucose metabolism [6]. This patient, however, did not present cerebellar features, but focal dystonia [6]. The significance of DJ-1 expression in the cerebellum still needs further investigation and it may be linked with other symptoms than PD symptoms. We also found a strong DJ-1 expression in non-motor structures like the hippocampus and the amygdala. A recent study revealed that PD patients with or without cognitive impairment displayed more often hippocampus atrophy than healthy subjects [5]. On the other hand, Dekker et al. did not report any dementia linked with DJ-1 mutation in PD patients [6]. These patients rather had psychiatric symptoms such as anxiety, paranoid delusions, and psychotic episodes. Since the amygdala and the hippocampus are key structures of the limbic system, the mutated DJ-1 in these structures may play a role in the above mentioned psychiatric symptoms. In the olfactory tubercle, we found a strong expression of DJ-1. Smell sense in patients with DJ-1 mutation is not known, but most patients with PD present a profound disorder of olfaction [9]. This might indicate that DJ-1 is involved in olfaction. Dysfunction of this protein may play an important role in the lack of smell not only in patients with a DJ-1 mutation but also in patients with other forms of PD. Hence, it would be interesting to investigate whether there is a difference of DJ-1 expression in the olfactory tubercle between PD patients and healthy subjects, and to investigate smell in PD patients with a DJ-1 mutation. The significance of DJ-1 expression in other brain regions of the brain such as the piriform cortex and cranial nerve nucleus is unclear. In summary, this study demonstrates a high expression of DJ-1 mRNA within neuronal and non-neuronal cells in the basal ganglia, thalamus, and cortical circuits. This finding suggests that mutation of DJ-1 could affect several motor structures known to play an important role in the development of PD. In addition, our findings of a strong expression of DJ-1 mRNA in the cerebellum, the hippocampus, and the olfactory tubercle indicates that a dysfunction of DJ-1 may contribute to additional non-motor clinical features of PD patients with a DJ-1 mutation.
H. Shang et al. / Neuroscience Letters 367 (2004) 273–277
Acknowledgements We thank Theres Lauterburg for technical assistance. We thank Christine Capper for valuable comments on the manuscript. This study was supported by a grant of the Foundation Telethon, Action Suisse (FTAS; to D.L.).
References [1] P.M. Abou-Sleiman, H.D., N. Quinn, A.J. Lees, N.W. Wood, The role of pathogenic DJ-1 mutations in Parkinson’s disease, Ann. Neurol. 54 (2003) 281–282. [2] R. Bandopadhyay, A.E. Kingsbury, M.R. Cookson, A.R. Reid, I.M. Evans, A.D. Hope, A.M. Pittman, T. Lashley, R. Canet-Aviles, D.W. Miller, C. McLendon, C. Strand, A.J. Leonard, P.M. Abou-Sleiman, D.G. Healy, H. Ariga, W.W. Wood, R. de Silva, T. Revesz, J.A. Hardy, J.A. Lees, The expression of DJ-1 (PARK7) in normal human CNS and idiopathic Parkinson’s disease, Brain 127 (2004) 420–430. [3] V. Bonifati, P. Rizzu, M.J. van Baren, O. Schaap, G.J. Breedveld, E. Krieger, M.C. Dekker, F. Squitieri, P. Ibanez, M. Joosse, J.W. van Dongen, N. Vanacore, J.C. van Swieten, A. Brice, G. Meco, C.M. van Duijn, B.A. Oostra, P. Heutink, Mutations in the DJ-1 gene associated with autosomal recessive early-onset parkinsonism, Science 299 (2003) 256–259. [4] J. Burgunder, A. Richter, W. Loscher, Expression of cholecystokinin, somatostatin, thyrotropin-releasing hormone, glutamic acid decarboxylase and tyrosine hydroxylase genes in the central nervous motor systems of the genetically dystonic hamster, Exp. Brain Res. 129 (1999) 114–120.
277
[5] R. Camicioli, M.M. Moore, A. Kinney, E. Corbridge, K. Glassberg, J.A. Kaye, Parkinson’s disease is associated with hippocampal atrophy, Mov. Disord. 18 (2003) 784–790. [6] M. Dekker, V. Bonifati, J. van Swieten, N. Leenders, R.J. Galjaard, P. Snijders, M. Horstink, P. Heutink, B. Oostra, C. van Duijn, Clinical features and neuroimaging of PARK7-linked parkinsonism, Mov. Disord. 18 (2003) 751–757. [7] B.I. Giasson, H. Ischiropoulos, V.M. Lee, J.Q. Trojanowski, The relationship between oxidative/nitrative stress and pathological inclusions in Alzheimer’s and Parkinson’s diseases, Free Radic. Biol. Med. 32 (2002) 1264–1275. [8] S. Hague, E. Rogaeva, D. Hernandez, C. Gulick, A. Singleton, M. Hanson, J. Johnson, R. Weiser, M. Gallardo, B. Ravina, K. Gwinn-Hardy, A. Crawley, P.H. St George-Hyslop, A.E. Lang, P. Heutink, V. Bonifati, J. Hardy, A. Singleton, Early-onset Parkinson’s disease caused by a compound heterozygous DJ-1 mutation, Ann. Neurol. 54 (2003) 271–274. [9] C. Hawkes, Olfaction in neurodegenerative disorder, Mov. Disord. 18 (2003) 364–372. [10] K. Honbou, N.N. Suzuki, M. Horiuchi, T. Niki, T. Taira, H. Ariga, F. Inagaki, The crystal structure of DJ-1, a protein related to male fertility and Parkinson’s disease, PG-31380-31384, J. Biol. Chem. 278 (2003) 31380–31384. [11] Q. Huai, Y. Sun, H. Wang, L.S. Chin, L. Li, H. Robinson, H. Ke, Crystal structure of DJ-1/RS and implication on familial Parkinson’s disease, FEBS Lett. 549 (2003) 171–175. [12] D. Nagakubo, T. Taira, H. Kitaura, M. Ikeda, K. Tamai, S.M. Iguchi-Ariga, H. Ariga, DJ-1, a novel oncogene which transforms mouse NIH3T3 cells in cooperation with ras, Biochem. Biophys. Res. Commun. 231 (1997) 509–513. [13] G. Paxinos, K.B.J. Franklin, The Mouse Brain in Stereotaxic Coordinates, Academic, Orlando, FL, 2001.
Localization of DJ-1 mRNA in the mouse brain Huifang Shang a,b , Doris Lang c , Burgunder Jean-Marc b,d , Alain Kaelin-Lang b,∗ a Department of Neurology, West China Hospital, SiChuan University, 610041 Chengdu, SiChuan, China Department of Neurology, Laboratory of Neuromorphology, Inselspittal, University of Bern, CH-3010 Bern, Switzerland c Neuro-Oncology Program, Division of Oncology, University Children’s Hospital Zurich, CH-8032 Zurich, Switzerland d Department of Medicine, National University of Singapore, 5 Lower Kent Ridge Road, Singapore 119074, Singapore
b
Received 13 February 2004; received in revised form 1 June 2004; accepted 2 June 2004
Abstract DJ-1 is mutated in autosomal recessive, early onset Parkinson’s disease but the exact localization of the DJ-1 gene product in the mammalian brain is largely unknown. We aimed to evaluate the DJ-1 mRNA expression pattern in the mouse brain. Serial coronal sections of brains of five male and five female adult mice were investigated by using in situ hybridization with a DJ-1 specific 35 S-labeled oligonucleotide probe. Hybridized sections were analyzed after exposure to autoradiography films and after coating with a photographic emulsion. DJ-1 was heterogeneously expressed throughout the mouse central nervous system. A high expression of DJ-1 mRNA was detected in neuronal and non-neuronal populations of several structures of the motor system such as the substantia nigra, the red nucleus, the caudate putamen, the globus pallidus, and the deep nuclei of the cerebellum. Furthermore, DJ-1 mRNA was also highly expressed in non-motor structures including the hippocampus, the olfactory bulb, the reticular nucleus of the thalamus, and the piriform cortex. The high expression of DJ-1 mRNA in brain regions involved in motor control is compatible with the occurrence of parkinsonian symptoms after DJ-1 mutations. However, expression in other regions indicates that a dysfunction of DJ-1 may contribute to additional clinical features in patients with a DJ-1 mutation. © 2004 Elsevier Ireland Ltd. All rights reserved. Theme: Motor systems and sensorimotor integration Topic: Basal ganglia, cortex, cerebellum Keywords: DJ-1 gene; Parkinson’s disease; In situ hybridization; Central nervous system
The DJ-1 gene was first identified as an oncogene [12] and contains eight exons spanning 24-kb. This gene encodes a highly conserved protein (DJ-1). Recently, mutations in the DJ-1 gene have been found to be associated with autosomal recessive, early onset Parkinson’s disease (PD) [3,8]. A 14-kb homozygous deletion in the DJ-1 gene, which removes exons 1 through 5, was identified to segregate with PD in a large Dutch family [3], and a homozygous 497T > C transition, which results in a missense mutation (L66P), was detected to segregate with PD in an Italian family [3]. Two other point mutations, R98Q and A104T, were identified in PD patients [8]. Two mutations (IVS6-1G → C and c.56delG c.57G → A), which were predicted to drastically change the protein product and result in the dysfunction of ∗ Corresponding author. Tel.: +41-31-632-21-11; fax: +41-31-632-27-52. E-mail address: [email protected] (A. Kaelin-Lang).
the protein, were also reported [8]. The mechanism by which altered DJ-1 induces PD is not known so far but several hypotheses have been presented [2]. Altogether, DJ-1 mutations account for a small number of autosomal recessive juvenile Parkinson cases [1]. In the human brain, northern blot studies revealed a ubiquitous expression of the DJ-1 gene particularly in subcortical structures, such as the caudate nucleus, the hippocampus, the thalamus, and the substantia nigra [3]. Using immunohistochemistry, abundant DJ-1 protein immunoreactivity was found in the frontal cortex and in the substantia nigra of normal, PD and PSP brain, and was mainly located in astrocytes [2]. However, the exact distribution and cellular localization of the DJ-1 mRNA, especially in other regions of the central nervous system is still unknown. The aim of the present study was to map the expression of the DJ-1 gene in the mouse brain by using in
0304-3940/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2004.06.002
274
H. Shang et al. / Neuroscience Letters 367 (2004) 273–277
situ hybridization (ISH) histochemistry with a 35 S-labeled oligonucleotide probe. Five male and five female adult mice were housed in a light- and temperature-controlled room (12-h dark:12-h light cycles) with normal mice chow and water ad libitum. These mice were decapitated under ether anesthesia. Their brains were then frozen in isopentane (at −40 ◦ C) and stored at −70 ◦ C. Serial coronal sections (12 m) of the entire brain were cut on a cryostat-microtome (Reichert-Jung, Germany) at −20 ◦ C and mounted onto gelatin-coated slides. The oligonucleotide probe for mouse DJ-1 mRNA was designed using the software Primer Designer 4 (Sci-ed software, Durham, USA; GenBank No.: AB015652). The sequence of the DJ-1 probe (anti-sense) is 5 ACAATGGCTAGTGCAAACTCAAAGCTGGTCCCCGGCCCGC-3 , which is complementary to the nucleotides 569–608 of mouse DJ-1 mRNA. As a control, a sense probe of the same nucleotides was used. The probes were commercially synthesized (Microsynth AG, Balgach, Switzerland). They were labeled with 35 S-deoxyadenosine-alpha-thiotriphosphate (35 S-␣-dATP, Hartmann Analytic, Braunschweig, Germany) using terminal deoxynucleotidyl transferase (Roche Diagnostics GmbH, Mannheim, Germany) according to the standard 3 end labeling reaction protocol of the manufacturer. The labeling reaction was stopped on ice and with 400 l TE buffer (0.1 M Tris–HCl pH 7.5, 10 mM EDTA), extracted and purified by phenol and chloroform, precipitated by ethanol and sodium chloride (NaCl) as described elsewhere [4]. Hybridization histochemistry was performed as previously described [4]. In brief, the sections were fixed with 4% formaldehyde in phosphate-buffered saline (PBS, pH 7.4) for 6 min. After washing twice with PBS, they were placed in 0.25% acetic anhydride in 0.1 M triethanolamine/0.9% NaCl (pH 8), for 10 min and dehydrated through a graded series of ethanol (70–100%). Sections were immersed in chloroform followed by a partial rehydration in ethanol (100–95%), and air dried at room temperature (RT). The slides were incubated with 35 S-labeled probes in 50 l of hybridization buffer containing 50% formamide (v/v), 600 mM NaCl, 80 mM Tris–HCl (pH 7.5), 4 mM EDTA, 0.1% (w/v) sodium pyrophosphate, 0.2% (w/v) SDS, 50 IU/ml Heparin, and 10% (w/v) Na-dextran sulfate at 37 ◦ C for 20 h. Posthybridization washing was performed in 0.5× SSC (1× SSC = 0.15 M NaCl/0.015 M sodium citrate) at 55 ◦ C for 4 × 15 min and in 0.5× SSC at RT for 2 × 30 min. The slides were air dried and exposed to Kodak BioMax MRI film (Eastman Kodak Company, Rochester, New York, USA) for 7 days. The autoradiography films were analyzed using the NIH Image software. The slides were then immersed in Ilford K5 nuclear emulsion (Ilford, Mobberly, Cheshire, UK) diluted in water (2:1, v/v) and exposed for 4 weeks at 4 ◦ C in darkness before being developed with Kodak D-19 for 4 min at 20 ◦ C, then fixed in Ilford Hypam fixer (Ilford; 1:4, v/v) for 4 min. After washing, the sections were counterstained with 0.2% toluidine
blue. The DJ-1 mRNA expression of the ISH experiments was evaluated according to silver grain densities on the sections. Signals were assessed as strongly positive, moderate, weakly positive, and not detectable and the localization was identified according to a standard mouse brain atlas [13]. Specificity of the probe and the hybridization procedure were checked as follows: (1) incubation with the 35 S-labeled sense probe; (2) pretreatment with ribronuclease (RNase) containing 20 g/ml RNase A (Sigma-Aldrich Corp., St. Louis, MO, USA) and 1 unit/ml RNase T1 (Sigma); (3) co-incubation with labeled and a 100-fold excess of unlabeled DJ-1 probe; and (4) co-incubation with a mixture of sense and anti-sense DJ-1 probe. Our results showed a similar DJ-1 mRNA expression pattern in all ten brains. The examination of the autoradiography films hybridized with the anti-sense DJ-1 oligonucleotide probe displayed a widespread expression (Fig. 1A–E). At the cellular level, examination of emulsion-coated tissue sections revealed a high number of specific silver grain clusters overlying toluidine blue counterstained cells with a typical neuronal morphology. No signal was seen on the film (Fig. 1G–I) and on emulsion-coated tissue sections processed with RNase pretreatment, sense DJ-1 probe, co-incubation with labeled and a 100-fold excess of unlabeled DJ-1 probe, and co-incubation with a mixture of sense and anti-sense DJ-1 probe. In both the olfactory bulb (Fig. 1F) and the lobules of cerebellum (Fig. 1J), a weak signal was seen with the sense probe. However this signal was much lower than when the antisense probe was used. Several structures involved in the regulation of motor activity had a high level of expression: neurons in the motor cortex (Fig. 2A), the caudate putamen (Fig. 2B), the substantia nigra (Fig. 2C), and the red nucleus (Fig. 2D) were strongly labeled. The cerebellum also exhibited strong DJ-1 expression (Fig. 1E), particularly in neurons of the granule cell layers and the deep cerebellar nuclei (Fig. 2E), and in Purkinje cells (data not shown). In contrast, the molecular cell layer in the cerebellum was only weakly labeled (data not shown). In addition, we also found several strongly labeled large cells in the lateral globus pallidus (data not shown) and other nuclei such as trigeminal motor neurons (Fig. 2F) in the brain stem. Other structures were also strongly labeled. In particular, strong hybridization signals were detected in the hippocampus (Fig. 1C and D), the piriform cortex (Fig. 1B), the olfactory cortex (Fig. 1A), and the reticular nucleus of the thalamus (Fig. 1C). By contrast, the cells in the secondary somatosensory cortex (data not shown) were only weakly labeled. In the brain structures with a high DJ-1 expression level and a high neuronal density, like the hippocampus, the olfactory bulb, and the cerebellum, the morphology of individual DJ-1 positive cells could not be assessed accurately. In all other structures, most of the DJ-1 positive cells had a typical neuronal morphology. A weaker more diffuse but specific signal was also observed between these neuronal cells
H. Shang et al. / Neuroscience Letters 367 (2004) 273–277
275
Fig. 1. Film autoradiogram showing the distribution of the DJ-1 mRNA in the adult mouse brain. Coronal sections hybridized with the radiolabeled oligonucleotide anti-sense probe (A–E) and sense probe (F–J) for DJ-1. Regions with high DJ-1 mRNA expression include olfactory bulb (A), caudate putamen (Cpu; B and C), piriform cortex (Pir; B), hippocampus (Hippo; C), lateral globus pallidus (LGP; C), medial globus pallidus (MGP; C), reticular thalamic nucleus (Rt; C), red nucleus (RN; D), substantia nigra (SN; D), amygdalohippocampal area (AHi; D), cerebellar cortex and dentate nucleus (DCN; E). Scale bar, 1 mm.
276
H. Shang et al. / Neuroscience Letters 367 (2004) 273–277
Fig. 2. Photomicrographs of emulsion-dipped sections depicting the cellular mRNA expression of DJ-1 in neurons of the mouse brain (bright field microscopy, counterstained with toluidine blue). (A) Motor cortex, (B) caudate putamen nucleus, (C) substantia nigra, (D) red nucleus, (E) dentate nucleus of the cerebellum, and (F) trigeminal motor nucleus. Scale bars, 100 m.
(Fig. 2). In contrast, no specific signal was detected on the surrounding white matter. In addition, other non-neuronal structures like the choroid plexus also exhibited DJ-1 expression. An important finding of our study is the demonstration of a widespread and strong DJ-1 expression in neuronal populations of several structures of the motor system. The high DJ-1 expression in neurons of the caudate putamen and substantia nigra of the mouse brain confirms previous results in human brain homogenates [3]. Since basal ganglia, thalamus, and motor cortex are structures involved in the pathophysiology of PD in humans, we can speculate that mutated DJ-1 could predominantly impair the function of these structures and thus lead to the development of PD symptoms. However, further studies are needed to investigate possible species difference in the pattern of DJ-1 gene expression. A recent study using immunohistochemistry on human brains demonstrated a stronger DJ-1 immunoreactivity in astrocytes than in neurons [2]. Also in our study we found a specific DJ-1 expression between neuronal cells that could be due to glial expression. However, further studies are needed to quantitatively investigate the level of expression as well as the subcellular localization of DJ-1 expression in neurons and glial cells. Although the mechanism by which mutated DJ-1 in neurons or astrocytes causes PD is unclear [2,3,10], recent studies show that DJ-1 may be involved in the cellular response to oxidative stress. The mutant DJ-1 protein may result in conformational changes of DJ-1, causing the mislocalization of DJ-1 to mitochondria or decrease the interaction of DJ-1 with the binding partners, and finally
decrease the ability to limit oxidative damage [3,10,11]. Since oxidative damage has been found to be involved in death of the nigral dopaminergic neurons and in the pathogenesis of PD [7], it is conceivable that the mutation of DJ-1 leads to the development of PD through oxidative stress. Furthermore, we found a strong DJ-1 expression in neurons of the cerebellum. This finding confirms and extends the previous report of moderate DJ-1 gene expression in tissue homogenates of the human cerebellum [3]. The clinical importance of this finding is unclear. Cerebellar symptoms have not been described in PD patients with DJ-1 mutation. However, a recent clinical study showed a possible cerebellar involvement in DJ-1 linked PD by assessing cerebral glucose metabolism [6]. This patient, however, did not present cerebellar features, but focal dystonia [6]. The significance of DJ-1 expression in the cerebellum still needs further investigation and it may be linked with other symptoms than PD symptoms. We also found a strong DJ-1 expression in non-motor structures like the hippocampus and the amygdala. A recent study revealed that PD patients with or without cognitive impairment displayed more often hippocampus atrophy than healthy subjects [5]. On the other hand, Dekker et al. did not report any dementia linked with DJ-1 mutation in PD patients [6]. These patients rather had psychiatric symptoms such as anxiety, paranoid delusions, and psychotic episodes. Since the amygdala and the hippocampus are key structures of the limbic system, the mutated DJ-1 in these structures may play a role in the above mentioned psychiatric symptoms. In the olfactory tubercle, we found a strong expression of DJ-1. Smell sense in patients with DJ-1 mutation is not known, but most patients with PD present a profound disorder of olfaction [9]. This might indicate that DJ-1 is involved in olfaction. Dysfunction of this protein may play an important role in the lack of smell not only in patients with a DJ-1 mutation but also in patients with other forms of PD. Hence, it would be interesting to investigate whether there is a difference of DJ-1 expression in the olfactory tubercle between PD patients and healthy subjects, and to investigate smell in PD patients with a DJ-1 mutation. The significance of DJ-1 expression in other brain regions of the brain such as the piriform cortex and cranial nerve nucleus is unclear. In summary, this study demonstrates a high expression of DJ-1 mRNA within neuronal and non-neuronal cells in the basal ganglia, thalamus, and cortical circuits. This finding suggests that mutation of DJ-1 could affect several motor structures known to play an important role in the development of PD. In addition, our findings of a strong expression of DJ-1 mRNA in the cerebellum, the hippocampus, and the olfactory tubercle indicates that a dysfunction of DJ-1 may contribute to additional non-motor clinical features of PD patients with a DJ-1 mutation.
H. Shang et al. / Neuroscience Letters 367 (2004) 273–277
Acknowledgements We thank Theres Lauterburg for technical assistance. We thank Christine Capper for valuable comments on the manuscript. This study was supported by a grant of the Foundation Telethon, Action Suisse (FTAS; to D.L.).
References [1] P.M. Abou-Sleiman, H.D., N. Quinn, A.J. Lees, N.W. Wood, The role of pathogenic DJ-1 mutations in Parkinson’s disease, Ann. Neurol. 54 (2003) 281–282. [2] R. Bandopadhyay, A.E. Kingsbury, M.R. Cookson, A.R. Reid, I.M. Evans, A.D. Hope, A.M. Pittman, T. Lashley, R. Canet-Aviles, D.W. Miller, C. McLendon, C. Strand, A.J. Leonard, P.M. Abou-Sleiman, D.G. Healy, H. Ariga, W.W. Wood, R. de Silva, T. Revesz, J.A. Hardy, J.A. Lees, The expression of DJ-1 (PARK7) in normal human CNS and idiopathic Parkinson’s disease, Brain 127 (2004) 420–430. [3] V. Bonifati, P. Rizzu, M.J. van Baren, O. Schaap, G.J. Breedveld, E. Krieger, M.C. Dekker, F. Squitieri, P. Ibanez, M. Joosse, J.W. van Dongen, N. Vanacore, J.C. van Swieten, A. Brice, G. Meco, C.M. van Duijn, B.A. Oostra, P. Heutink, Mutations in the DJ-1 gene associated with autosomal recessive early-onset parkinsonism, Science 299 (2003) 256–259. [4] J. Burgunder, A. Richter, W. Loscher, Expression of cholecystokinin, somatostatin, thyrotropin-releasing hormone, glutamic acid decarboxylase and tyrosine hydroxylase genes in the central nervous motor systems of the genetically dystonic hamster, Exp. Brain Res. 129 (1999) 114–120.
277
[5] R. Camicioli, M.M. Moore, A. Kinney, E. Corbridge, K. Glassberg, J.A. Kaye, Parkinson’s disease is associated with hippocampal atrophy, Mov. Disord. 18 (2003) 784–790. [6] M. Dekker, V. Bonifati, J. van Swieten, N. Leenders, R.J. Galjaard, P. Snijders, M. Horstink, P. Heutink, B. Oostra, C. van Duijn, Clinical features and neuroimaging of PARK7-linked parkinsonism, Mov. Disord. 18 (2003) 751–757. [7] B.I. Giasson, H. Ischiropoulos, V.M. Lee, J.Q. Trojanowski, The relationship between oxidative/nitrative stress and pathological inclusions in Alzheimer’s and Parkinson’s diseases, Free Radic. Biol. Med. 32 (2002) 1264–1275. [8] S. Hague, E. Rogaeva, D. Hernandez, C. Gulick, A. Singleton, M. Hanson, J. Johnson, R. Weiser, M. Gallardo, B. Ravina, K. Gwinn-Hardy, A. Crawley, P.H. St George-Hyslop, A.E. Lang, P. Heutink, V. Bonifati, J. Hardy, A. Singleton, Early-onset Parkinson’s disease caused by a compound heterozygous DJ-1 mutation, Ann. Neurol. 54 (2003) 271–274. [9] C. Hawkes, Olfaction in neurodegenerative disorder, Mov. Disord. 18 (2003) 364–372. [10] K. Honbou, N.N. Suzuki, M. Horiuchi, T. Niki, T. Taira, H. Ariga, F. Inagaki, The crystal structure of DJ-1, a protein related to male fertility and Parkinson’s disease, PG-31380-31384, J. Biol. Chem. 278 (2003) 31380–31384. [11] Q. Huai, Y. Sun, H. Wang, L.S. Chin, L. Li, H. Robinson, H. Ke, Crystal structure of DJ-1/RS and implication on familial Parkinson’s disease, FEBS Lett. 549 (2003) 171–175. [12] D. Nagakubo, T. Taira, H. Kitaura, M. Ikeda, K. Tamai, S.M. Iguchi-Ariga, H. Ariga, DJ-1, a novel oncogene which transforms mouse NIH3T3 cells in cooperation with ras, Biochem. Biophys. Res. Commun. 231 (1997) 509–513. [13] G. Paxinos, K.B.J. Franklin, The Mouse Brain in Stereotaxic Coordinates, Academic, Orlando, FL, 2001.