Huntingtin-interacting protein-1-related protein of rat (rHIP1R) is localized in the postsynaptic regions

Huntingtin-interacting protein-1-related protein of rat (rHIP1R) is localized in the postsynaptic regions

Brain Research 967 (2003) 210–225 www.elsevier.com / locate / brainres Research report Huntingtin-interacting protein-1-related protein of rat (rHIP...

2MB Sizes 0 Downloads 3 Views

Brain Research 967 (2003) 210–225 www.elsevier.com / locate / brainres

Research report

Huntingtin-interacting protein-1-related protein of rat (rHIP1R) is localized in the postsynaptic regions Akira Okano a , Nobuteru Usuda b , Kenichi Furihata c , Kouzoh Nakayama d , Qing Bao Tian a , Takashi Okamoto e , Tatsuo Suzuki a , * a

Department of Neuroplasticity, Research Center on Aging and Adaptation, University School of Medicine, 3 -1 -1 Asahi, Matsumoto 390 -8621, Japan b Department of Anatomy, Fujita Health Science University School of Medicine, Toyoake, Aich 470 -1192, Japan c Department of Molecular and Experimental Medicine, Scripps Research Institute, Maildrop MEM-150, 10550 North Torrey Pines Road, La Jolla, CA 92037, USA d Department of Anatomy, Shinshu University School of Medicine, 3 -1 -1 Asahi, Matsumoto 390 -8621, Japan e Department of Molecular Genetics, Nagoya City University Medical School, 1 Kawasumi, Mizuho-ku, Nagoya 467, Japan Accepted 3 January 2003

Abstract We cloned a rHIP1R (GenBank Accession No., AB005052) encoding a Sla2 / huntingtin-interacting protein (HIP1) family protein from a rat brain cDNA library. Localization of rHIP1R was investigated in the rat brain using an antibody specific to the HIP1R antibody. The rHIP1R protein was enriched in the synaptic plasma membrane fraction along with huntingtin, a synaptic protein and a causal protein for Huntington’s disease. The electron microscopic examination revealed that HIP1R was localized at postsynaptic spines. Localization of HIP1R in the small vesicular structures in the spine, possible sites of vesicular transport of synaptic proteins, together with the structure-based analysis, suggested a role of HIP1R for vesicle trafficking through interaction with F-actin and working together with huntingtin and HIP1 at the synaptic sites.  2003 Elsevier Science B.V. All rights reserved. Theme: Excitable membranes and synaptic transmission Topic: Postsynaptic mechanisms Keywords: Postsynaptic; Huntingtin; Membrane trafficking

1. Introduction Huntingtin, a causal protein for Huntington’s disease, is localized in neurons throughout the brain, and is primarily found in somatodendritic regions [9]. Huntingtin is localized in the synaptic region and is suggested to have a role Abbreviations: CHAPS, 3-[(3-cholamidopropyl)dimethyl-ammonino]1-propanesulfonate; DED, death effector domain; DIG, digoxigenin; EH, Eps15 homology; ENTH, epsin N-terminal homology; ER, endoplasmic reticulum; GST, glutathione S-transferase; HAP, huntingtin-associated protein; HIP, huntingtin-interacting protein; IPTG, isopropyl-b-Dthiogalactopyranoside; PCR, polymerase chain reaction; RACE, rapid amplification of cDNA ends; RT-PCR, reverse transcriptase-based PCR; SPM, synaptic plasma membrane *Corresponding author. Tel.: 181-263-37-2683; fax: 181-263-372725. E-mail address: [email protected] (T. Suzuki).

in the synaptic function [5,9]. Huntingtin colocalizes with microtubules and is suggested to be involved in vesicle trafficking [1,5,33,37]. Various huntingtin-interacting proteins (HIPs) have been identified, including HIP1 [12,35], HIP2 (or hE2-25K) [12], huntingtin-associated protein 1 (HAP1) [15], and glyceraldehyde 3-phosphate dehydrogenase [2]. Human HIP1 (hHIP1) is predominantly expressed in the brain, co-localizes with huntingtin, and interacts with the Nterminal portion of huntingtin [13]. The binding affinity of HIP1 to huntingtin is inversely correlated to the huntingtin polyglutamine length. Loss or reduction of huntingtinHIP1 interaction due to the extension of CAG repeats in the huntingtin gene is considered attributable to defective membrane-cytoskeletal integrity in the brain, contributing to the development of Huntington’s disease. In addition, the appearance of a free form of HIP1 is considered to

0006-8993 / 03 / $ – see front matter  2003 Elsevier Science B.V. All rights reserved. doi:10.1016 / S0006-8993(03)02236-4

A. Okano et al. / Brain Research 967 (2003) 210–225

cause apoptosis of neurons and culminate in Huntington’s disease [10]. HAP1 was cloned by yeast two-hybrid screening using a huntingtin cDNA fragment containing 44 polyglutamine repeats as bait [15]. The affinity of the HAP1 to huntingtin increases dependently on the length of the polyglutamine repeat. HIP1 shares the characteristic structure with Sla2. Sla2, also known as Mop2 or End4, is an actin-binding protein and is involved in the actin organization and endocytosis in yeast [11]. Sla2 appears to be conserved from yeast to mammals. Homologs of Sla2p have been identified in nematodes (ZK370.3) [36], humans (hHIP1, hHIP1R or hHIP12) [4,13,25] and mice (mHIP1R or mHIP12) [6]. The C-terminus of these proteins, containing the I / LWEQ module (an F-actin-binding domain) [18], is highly homologous to talin, a membrane-cytoskeleton linker protein [3]. The Sla2 / HIP1 family proteins also possess the epsin N-terminal homology (ENTH) domain at their N-termini. A number of ENTH-containing proteins were suggested to interact with clathrin and to be involved in endocytosis and / or regulation of cytoskeletal organization [14]. Thus, the Sla2 / HIP1 family proteins could be involved in vesicle trafficking through interaction with actin-based cytoskeletons. In the present study, we isolated a cDNA clone of HIP1-related protein (HIP1R) from the rat cDNA library and compared the structure of the encoding protein with those from mice and human. We then produced an affinitypurified anti-rat HIP1R (rHIP1R) antibody and demonstrated evidence of synaptic localization of the HIP1R, suggesting a role for HIP1R at the synapse.

211

blotting were from Pierce (Rockford, IL); All other chemicals were of reagent grade.

2.2. Cloning of rHIP1 R Clone M11 with an insert size of 2510 bp was isolated from the lZAP II cDNA library prepared from the rat brain cortex in a survey of genes encoding synaptic proteins [16]. We performed cloning of a full-length cDNA, because clone M11 had not been previously reported in the initial database. Molecular cloning was carried out according to the manual for the picoBlue Immunoscreening kit. Then, clones M11-1377 bp and M11-842 bp were obtained from the same library. The cDNA clone rHIP1R containing the largest insert of 3646 bp was isolated by screening using M11-842 bp as a probe that was random prime labeled with digoxigenin (DIG). The remaining 59 portion of the rHIP1R cDNA was isolated using 59 RACE. The first strand cDNA was synthesized with the oligo-dT. After dATP tailing of the first-strand cDNA with terminal deoxynucleotide transferase, 35 cycles (94 8C for 30 s; 55 8C for 30 s; 72 8C for 1 min) of PCR using rHIP1R complementary sequence (residues 941–922, ACCGCCTTGATGTGCTCAGC) and the oligo-dT anchor primer was carried out. The primers used for the second round of PCR (94 8C for 30 s; 60 8C for 30 s; 72 8C for 1 min) were a nested primer (rHIP1R complementary sequence: residues 875–854, GATCTGGATGAGTCGCTTG) and a supplied anchor primer. The 0.8-kbp PCR product was cloned using the pGEM-T Easy Vector System and merged to the 3646-bp rHIP1R cDNA fragment. The GenBank accession number for the final rHIP1R clone is AB005052.

2. Materials and methods

2.1. Materials

2.3. mRNA isolation, reverse transcriptase-based PCR ( RT-PCR) and Northern blotting

The cDNA Lambda ZAP II library from rat brain cortex, a picoBlue immunoscreening kit and Bluescript plasmid were purchased from Stratagene (La Jolla, CA). pGEM-T easy vector was from Promega Corporation (Madison, WI); pGEX-4T-1 vector was from Amersham–Pharmacia Biotech (Tokyo); Isogen from Nippon gene KK (Tokyo); SuperScript II preamplification’s system was from GibcoBRL; 59 / 39 Rapid Amplification of cDNA Ends (RACE) kit was from Boehringer-Mannheim Biochemica (Tokyo); RNA polymerase chain reaction (PCR) kit (avian myeloblastosis virus) Ver 2.1 was from Takara (Ohtsu, Japan); rabbit polyclonal anti-actin antibody (A2066), protease inhibitor cocktail, anti-rabbit IgG-agarose (A-1027), and anti-mouse IgG-agarose beads (A-6531) were from Sigma (St. Louis, MO); mouse monoclonal anti-huntingtin antibody (MAB2174) was from Chemicon International (Temecula, CA); anti-glutathione S-transferase (GST) antibody was from Amersham–Pharmacia Biotech (Bucks, UK); chemiluminescent detection reagents for Western

Total RNA was isolated from the rat forebrain using Isogen according to the manufacturer’s instructions. The RNA was incubated with DNase I at 37 8C for 30 min, then treated by Isogen again. RT-PCR was carried out according to the protocol of the RNA PCR kit Ver 2.1 with 1 mg of total RNA. The cDNA was synthesized at 42 8C for 30 min, followed by PCR (94 8C for 30 s, 60 8C for 30 s, and 72 8C for 1 min). Then, the PCR products were separated by electrophoresis using 1% agarose gel and visualized by ethidium bromide. For Northern blotting, mRNA (2 mg) was size-fractionated on formaldehyde-containing 1% agarose gels and capillary-blotted to nylon membranes. The filter was hybridized with random-primed 32 P-labeled probe that was generated using rHIP1R cDNAs (nt. 648–3291) as a template. After hybridization at 70 8C, the membrane was washed at high stringency at 70 8C with 0.13 SSC and 0.5% SDS, and exposed to imaging plate (Fuji Xerox, Tokyo).

212

A. Okano et al. / Brain Research 967 (2003) 210–225

2.4. Preparation of antibodies against rHIP1 R protein GST-rHIP1R C-terminal fusion protein (amino acid residuse 282–1079; predicted molecular weight, 88 342) was expressed in E. coli BL21 cells and purified by glutathione–Sepharose 4B using the GST gene fusion system of Amersham–Pharmacia Biotech. In brief, the recombinant plasmids (pGEX-4T-1 vector) were grown in protease-deficient E. coli cultures at 37 8C and transcription was induced by 50 mM isopropyl-b-D-thiogalactopyranoside (IPTG) for 1.5 h at 37 8C. The cells were pelletted and resuspended in 13 phosphate-buffered saline, then added 10 mM dithiothreitol. Cell lysates obtained after sonication with Vibro Cell (Sonics and Materials, Danbury, CT) were treated with 2% Triton X-100 and centrifuged at 100 0003g for 1 h. The supernatant obtained was applied to a glutathione–Sepharose 4B column and the fusion protein was eluted with thrombin. The eluted proteins were checked by its amino acid sequence. The protein was further purified by SDS–PAGE and the rHIP1R protein bands were isolated and used for immunogen. Polyclonal anti-rHIP1R antibody was raised in a rabbit and affinity-purified using GST-rHIP1R C-terminal fusion protein immobilized to Affi-Gel 10. Affinitypurified anti-rHIP1R antibody was used for Western blotting, immunoprecipitation and immunohistochemical studies, unless otherwise indicated.

100 mM KCl, 0.5 mM dithiothreitol, 0.2 mM ATP, 1 mM MgCl 2 , 0.2 mM CaCl 2 , and protease inhibitor cocktail). The mixtures were incubated at 28 8C for 4 h to polymerize actin and then centrifuged at 100 0003g for 1 h at 4 8C. Both supernatant and pellet were separated by SDS– PAGE and analyzed by Western blotting.

2.7. Immunohistochemical examination Brains from Wistar rats (6 weeks old) were processed essentially as described previously for immunohistochemical examination at the light microscopic and electron microscopic levels [31] using affinity-purified anti-rHIP1R antibody. Animals were handled in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals (NIH Publications No. 80-23). The 3,39-diaminobenzidine reaction product was developed in the presence of 0.025% cobalt chloride and 0.02% nickel ammonium to enhance the staining [22].

2.8. Subcellular fractionation Subcellular fractions of the forebrain were prepared from Wistar rats (6 weeks old, male) as described previously [22,32]. All preparations were stored at 280 8C until use.

2.5. Immunoprecipitation

3. Results

Proteins of synaptic plasma membrane (SPM) were solubilized for immunoprecipitation at 4 8C for 2 h in low stringent solubilizing buffer [29] containing 50 mM Tris– HCl (pH 7.5), 150 mM NaCl, 10 mM EDTA, 2 mM EGTA, 0.1% SDS, 1% Triton X-100, 1% 3-[(3cholamidopropyl)dimethyl-ammonino]-1-propanesulfonate (CHAPS), 0.5% Nonidet P-40, 0.1% bovine serum albumin, 50 mM NaF, 100 mM Na vanadate and protease inhibitor cocktail. Supernatant was obtained after centrifugation at 10 0003g for 10 min and incubated at 4 8C overnight with primary antibodies immobilized chemically to protein G plus / protein A-agarose. The gel was washed three times, and immunoprecipitated proteins were separated by SDS–PAGE and detected by Western blotting using chemiluminescent substrates [32]. Four hundred or 1500 mg of SPM protein were used for co-immunoprecipitation of rHIP1R for actin or huntingtin, respectively.

3.1. Cloning and sequence analysis of rHIP1 R cDNA

2.6. Co-sedimentation assay with F-actin Association with F-actin was examined according to the method by Nakamura et al. [23]. Actin prepared from rabbit skeletal muscle [28] was resuspended either with GST-rHIP1R C-terminal fusion protein or GST protein in actin polymerization buffer (10 mM Tris–HCl, pH 8.0,

The rHIP1R cDNA contained 4293 bases with a 3240base open reading frame, which encodes a 1079-amino acid polypeptide highly homologous to HIP1R of mice and humans (mHIP1R and HIP12, respectively). The calculated molecular weight and pI of the deduced rHIP1R protein were 120 565 and 6.11, respectively. A domain search using the SMART software program [24] revealed that the deduced rHIP1R protein had ENTH (amino acids 41–163) and I / LWEQ domains (amino acids 825–1023) in the N-terminus and C-terminus, respectively, as seen in other Sla2 / HIP family proteins (Fig. 1a). The coiled-coil structure was found in all HIP1 / Sla2 family proteins described above, but not in talin or epsin (not shown), which are homologous to but not members of HIP1 / Sla2 protein family based on analysis using MacSTrip software [17]. Homology of the death effecter domain (DED)-like domain in rat, mouse and human HIP1R proteins to hHIP1 DED was relatively low with 51.2% identity in 82 amino acid overlaps (Fig. 1b), and phenylalanine-398 in the HIP1, critical for the apoptosis-inducing activity [10], was replaced by glutamine in all HIP1Rs. The huntingtinbinding domain of HIP1 (amino acid resides 483–607 of hHIP1, GenBank Accession No. XP]004910) [4,13] was not conserved in in the rat, mouse and human HIP1R

A. Okano et al. / Brain Research 967 (2003) 210–225

213

Fig. 1. Domain structure of the rHIP1R protein and comparison of domain sequences between HIP1R and HIP. (a) Molecular structures were depicted based principally on SMART software (http: / / smart.embl-heidelberg.de) [24]. Leucine zipper (L-Z) analysis was performed using Scan Prosite (http: / / www.expasy.ch / tools / scnpsite.html). C-C refers to the coiled-coil structure. The length of polypeptides is shown in amino acid number on the top. (b) Multiple alignment of hHIP1 DED and its corresponding regions in rHIP1R, mHIP1R and hHIP1R. Asterisks and an arrowhead indicate amino acid residues that are highly conserved in DED of various DED-containing proteins and the position of the phenylalanine residue critical for the apoptosis-inducing activity, respectively [10]. (c) Multiple alignment of the huntingtin-binding region (Huntingtin-BD) of HIP1 and its corresponding regions in rHIP1R, mHIP1R and hHIP1R. Regions with three or four identical amino acid residues in these four proteins are boxed.

proteins (31.7% identity in the maximally matched a 161amino acid region) as shown in Fig. 1c. Interaction between rHIP1R and huntingtin was not detected by the co-immunoprecipitation method (not shown). Thus, HIP1R appears to be different from HIP1 at least in the DED and huntingtin-binding domains.

3.2. mRNA expression of rHIP1 R in rat brain Expression of rHIP1R mRNA in the brain was con-

firmed by both Northern blotting and RT-PCR (Fig. 2). Northern blotting showed a 4.2-kb single band (Fig. 2a). Expression of the gene in the rat brain was observed as early as 1 day after birth, and remained relatively unchanged in adulthood as demonstrated by RT-PCR (Fig. 2b). The same developmental expression pattern was obtained in the Northern blot analysis (not shown). This developmental expression pattern of rHIP1R resembled that of huntingtin gene in the rat brain [26]. Expression in the brain was also examined by in situ

214

A. Okano et al. / Brain Research 967 (2003) 210–225

Fig. 2. Expression of rHIP1R in the brain. (a) Northern blot of rHIP1R in the forebrain of adult rats. mRNA was prepared from the rat forebrain. (b) Developmental changes in the rHIP1R expression demonstrated by RTPCR. Total RNAs prepared from the rat forebrain at postnatal day 1 (1d) through 20 weeks (20w) were used. G3PDH refers to glyceraldehyde 3-phosphate dehydrogenase.

hybridization and it demonstrated ubiquitous expression of the rHIP1R gene in neurons throughout the adult brain, although the expression level appeared to be low judging from the signal intensity (Fig. 3). The signals, as seen in somas of the cortical neurons (Fig. 3a) and the hippocampal pyramidal neurons (Fig. 3b), were considered to be specific as in situ hybridization using the sense probe for the rHIP1R gene (used as a control) did not demonstrate any staining (Fig. 3c,d).

3.3. Production of a polyclonal antibody specific to HIP1 R protein A polyclonal antibody against rHIP1R protein was produced as described in Section 2 and Fig. 4 summarizes the process of antibody preparation. Lysate of E. coli BL21 cells expressing GST-rHIP1R C-terminal fusion protein (predicted molecular weight, 120 kDa) (lane 1) was loaded to the glutathione–Sepharose 4B column. The sepharose retained several bands including 120-kDa doublet bands (lane 2). Bands smaller than the 120-kDa doublet may be degraded or incompletely translated rHIP1R proteins. Microsequence of the doublet proteins demonstrated that the lower but not the upper band contained the sequence of rHIP1R. Thrombin treatment of the glutathione-eluate (lane 2) produced several bands (lane 3) and amino acid sequencing of the major bands numbered 1, 2, and 3 demonstrated that band 1 was HIP1R C-terminal peptide (predicted molecular weight, 88 342). Band 2 did not contain the rHIP1R sequence. Band 3 may be a degradation product or incompletely translated protein, containing the rHIP1R sequence. Thus, protein band 1 was cut, purified again by SDS–PAGE, and used as an immunogen.

Fig. 3. In situ hybridization of rHIP1R. Expression of rHIP1R in the cerebral cortex (a,c) and hippocampus (b,d) of an adult rat. An anti-sense probe was used for (a) and (b), and a sense probe for (c) and (d) as controls. O, Py and R refer to the stratum oriens, stratum pyramidale, stratum radiatum, respectively. Bar, 50 mm.

The antiserum obtained was checked by Western blotting as shown in Fig. 4b. The antiserum reacted to the translated GST-rHIP1R C-terminal fusion protein of 120 kDa and various degraded or incompletely translated GSTrHIP1R C-terminal proteins (lanes 4–7). The antiserum was also tested against the glutathione eluate (lane 8) and thrombin eluate obtained from the Sepharose 4B after the transfected cell lysate had been applied (lane 9). The antiserum reacted with bands 1 and 3 but not 2 in lane 9. The immunoreactive bands immediately beneath band 1, which appeared in a picture taken at higher sensitivity, may be degradation products of band 1. The antiserum did not react to GST (not shown). Thus, the antiserum was judged to be specific to the rHIP1R protein. The antirHIP1R antibody was further affinity-purified using GSTrHIP1R C-terminal fusion protein immobilized to Affi-Gel 10 and the affinity-purified antibody detected a single

A. Okano et al. / Brain Research 967 (2003) 210–225

215

Fig. 4. Preparation of anti-rHIP1R antibody. (a) Purification of rHIP1R C-terminal protein that was used for immunogen. Lane 1, extract of BL21 cells expressing GST-rHIP1R C-terminal fusion protein. Lane 2, glutathione eluate of glutathione–Sepharose 4B column applied with the transfected cell lysate. Lane 3, thrombin-treated glutathione eluate from the column. The bands produced were numbered bands 1, 2 and 3. All lanes were stained with Coomassie brilliant blue R-250. (b) Verification of the anti-rHIP1R antibody by Western blotting. Lanes 4–7 are lysates of BL21 cells in which translation of GST-rHIP1R C-terminal protein was either induced (lanes 5 and 7) or not (lanes 4 and 6) with IPTG. Lanes 4 and 5 are stained with Coomassie brilliant blue R-250, while those 6 and 7 are Western blotted with anti-rHIP1R anti-serum. Lane 8 is glutathione eluate of glutathione–Sepharose 4B after application of the transfected cells lysate. Lane 9 is thrombin eluate from the glutathione–Sepharose 4B column after application of BL21 cell extract. The band immediately below band 1 became dense and other faint bands in lane 9 became visible due to high sensitivity. The band numbers correspond to those in lane 3. Lanes 8 and 9 are Western blot using anti-rHIP1R anti-serum. Lanes 10 and 11, Western blotting of forebrain homogenate (200 mg protein) detected by affinity-purified anti-rHIP1R antibody, or preabsorbed antibody, respectively.

120-kDa protein band in the forebrain homogenate (lane 10). The band was eliminated when using affinity-purified antibody preabsorbed with the antigen used for antibody production (lane 11).

3.4. Interaction of rHIP1 R with F-actin and huntingtin The presence of C-terminal I / LWEQ domain suggested that rHIP1R was an actin-binding protein [18]. Therefore, binding of rHIP1R to actin filaments was investigated by spin-down experiments. Actin polymerization was induced either in the presence or absence of the affinity-purified GST-fusion protein with the C-terminal of rHIP1R protein containing the entire I / LWEQ domain. After actin polymerization and centrifugation, pelletted F-actin and the remnant supernatant was examined by Western blotting. As shown in Fig. 5a, GST-rHIP1R C-terminal protein cosedimented with F-actin. This co-sedimentation was considered to be specific because the GST-rHIP1R C-terminal alone, without actin, did not leave a sediment under the conditions used (Fig. 5a) and GST protein alone did not co-sediment F-actin (Fig. 5a). Interaction of rHIP1R with actin was also examined by co-immunoprecipitation assay. Actin was co-precipitated when SPM proteins were immunoprecipitated with antirHIP1R antibody (Fig. 5b). Omission of anti-rHIP1R antibody sedimented neither rHIP1R nor actin, thus the precipitation was considered to be specific.

3.5. Tissue and subcellular distributions of rHIP1 R protein The tissue distribution of rHIP1R protein was examined by Western blotting as shown in Fig. 6a. rHIP1R protein with the expected 120 kDa was detected in the brain, testis and spleen. Huntingtin protein was also detected in these tissues but not among the other tissues examined (Fig. 6b). Other immunoreactive bands with different molecular weights, especially the 97-kDa band, were detected in the stomach, heart, liver, skeletal muscle, intestine, spleen and pancreas but not in the brain. The relation of these proteins with the 120-kDa protein is currently unknown. The expression level of the 120-kDa band was relatively low judging from protein amount required for detection by Western blotting. The subcellular distribution of rHIP1R was also examined by Western blotting as shown in Fig. 7a. The rHIP1R immunoreactive protein of 120 kDa was detected in the total homogenate, P1 (cell debris and nuclei), the synaptosome and the SPM fractions. The rHIP1R-immunoreactive protein was enriched in the SPM fraction and a faint band was detected in the synaptosome fraction, but was not present in the PSD fraction that was obtained after Triton X-100 treatment and Triton X-100–KCl washing of SPM. Huntingtin also showed a similar distribution pattern in the synaptic fractions, but differing from the rHIP1R protein, huntingtin was also enriched in the soluble frac-

216

A. Okano et al. / Brain Research 967 (2003) 210–225

Fig. 5. Interaction of the rHIP1R protein and actin revealed by co-sedimentation assay and co-immunoprecipitation. (a) GST-rHIP1R fusion protein or GST alone was added to actin polymerization buffer and incubated in the presence or absence of actin. The supernatant (S) and pellets (P) obtained by centrifugation were separated by SDS–PAGE and Western blotted with anti-rHIP1R (1 / 200 dilution) or anti-GST antibody (1 / 4000 dilution). (b) SPM proteins (400 mg) were solubilized and immunoprecipitated with anti-rHIP1R antibody. Precipitated proteins were separated by SDS–PAGE and Western blotted with anti-b-actin (1 / 5000 dilution) or anti-rHIP1R (1 / 1000 dilution) antibody. For a control, SPM proteins were incubated in the absence of the antibody (marked as –). IP and WB refer to immunoprecipitation and Western blotting, respectively.

A. Okano et al. / Brain Research 967 (2003) 210–225

217

Fig. 6. Tissue distribution of rHIP1R. Proteins from various tissues (200 mg each) were separated on a 7% polyacrylamide gel and Western blotted using an affinity-purified anti-rHIP1R (1 / 1000 dilution) (a) and anti-huntingtin (1 / 1000 dilution) (b) antibody. Br, Sto, Hea, Kid, Liv, Thy, Tes, Mus, Int, Spl and Pan refer to brain, stomach, heart, kidney, liver, thymus, testis, skeletal muscle, intestine, spleen and pancreas, respectively. Arrows in (a) and (b) indicate rHIP1R and huntingtin, respectively. Molecular weights are indicated in kDa on the left.

218

A. Okano et al. / Brain Research 967 (2003) 210–225

Fig. 7. Subcellular distribution of rHIP1R protein in the rat brain and comparison with that of huntingtin. Proteins from various subcellular fractions (200 mg each) were separated on a 7% polyacrylamide gel and Western blotted using an affinity-purified anti-rHIP1R (1 / 1000 dilution) (a) and anti-huntingtin (1 / 1000 dilution) (b) antibody. H, S and Syn refer to whole forebrain homogenate, soluble fraction and synaptosome fraction, respectively. Arrows in (a) and (b) indicate rHIP1R and huntingtin, respectively. Molecular weights are indicated in kDa on the left.

tion (Fig. 7b). Synaptic localization of huntingtin was in good agreement with previous findings that huntingtin was detected predominantly in the neurons of the brain and, at the cellular level, in the dendritic spine [9]. There were

immunoreactive bands with different sizes, although very faint, detected in the supernatant, synaptosome and SPM fractions. The relation of these proteins with 120 kDa rHIP1R is unknown at present.

A. Okano et al. / Brain Research 967 (2003) 210–225

3.6. Immunohistochemical localization of rHIP1 R in the rat brain The rHIP1R immunoreactivity was widely distributed throughout the rat brain as observed in mRNA expression. Small immunoreactive dot-like signals were dispersed throughout the brain. The validity of immunohistochemical staining with affinity-purified anti-rHIP1R antibody was confirmed by the complete elimination of staining when the antibody was preabsorbed with the antigen used for the antibody production or without the primary antibody (data not shown). Typical examples of the distribution of rHIP1R immunoreactivity are shown in Fig. 8. In the neocortex, most pyramidal cells were positive for rHIP1R as shown in Fig.

219

8a. Immunoreactive small dot-like signals were found scattered throughout the tissues. The dot-like immunoreactive signals were detected in somas and dendrites of the neurons and, typically, they were distributed on long single dendrites as shown in Fig. 8b,c. In the hippocampus, somas of most pyramidal neurons in CA1–CA4 and dentate nucleus were positive for rHIP1R (Fig. 8d). A magnified picture shows the staining of the CA1 pyramidal neurons (Fig. 8e), where some of the nuclei appeared to be positive for the staining. Numerous small immunoreactive dots were found scattered in all hippocampal areas similar to in cerebral cortex. In the cerebellum, the molecular cell layers were darkly stained, and Purkinje cells and granule cells were also densely stained. In the molecular layer, densely stained

Fig. 8. Localization of rHIP1R in the rat brain at the light microscopic level. (a–c) Cerebral cortex. (d) Low magnification of hippocampus. (e) CA1 region of hippocampus. (f) White matter. (g–i) Cerebellum. Purkinje cell dendrites with immunoreactive dots are clearly shown in (h) and (i). Arrows in (b) and (h) show typical dendrites of cortical neurons and Purkinje cells, respectively. M, P and G in (g) refer to molecular layer, Purkinje cell layer and granular layer, respectively. Bars in (a) and (d), 200 mm. Bars in (b), (c), (e)–(i), 50 mm.

220

A. Okano et al. / Brain Research 967 (2003) 210–225

Fig. 8. (continued)

A. Okano et al. / Brain Research 967 (2003) 210–225

Purkinje cell dendrites were clearly stained by anti-rHIP1R antibody (Fig. 8h). Small immunoreactive dots were distributed on the somas of Purkinje cells and alongside the Purkinje cell dendrites (Fig. 8h,i). Cells in white matter, probably oligodendrocytes, were also stained (Fig. 8f). Dot-like staining in neural cells and synaptic distribution of the rHIP1R in the subcellular fractionation prompted us to examine the synaptic localization of this protein in the cerebral cortex using electron microscopy. We adopted an enhancement method of diaminobenzidine reaction products with nickel and cobalt [22] because the staining was rather weak in the light microscopic examination. Both specimens, with (Fig. 9) or without (Fig. 10) post-staining with lead citrate and uranyl acetate, were compared, because PSDs sometimes become highly electron-dense after the electron staining, and the structures visualized without poststaining could not easily be identified. Immunoreactivities were observed in the cytoplasmic structures in the spines including PSDs and structures which were attached or very close to PSD and apart from PSD (Fig. 9). Very small vesicular structures were observed in the cytoplasmic immunoreactive areas in the spine (open arrows in Fig. 9). Immunoreactivities were also found in the dendrites (arrows in Fig. 9c). The presynaptic terminals were constantly negative for staining. Mitochondrion in the synaptic terminals and dendrites were not stained. Both immunoreactive and non-reactive spines were observed. We confirmed the specificity of the immunostaining by its absence on sections incubated with the preabsorbed anti-rHIP1R antibody (Fig. 9a). The staining pattern was consistent with those in the sections without post-staining with lead citrate and uranyl acetate (Fig. 10), where the possibility of non-specific staining of PSD was ruled out. Immunoreactivities of PSDs were rather weak in these sections. Again small vesicular structures were observed in the immunoreactive cytoplasmic areas in the spine (open arrows in Fig. 10).

4. Discussion

221

Since the ENTH domain is involved in endocytosis and / or regulation of cytoskeletal organization [14] and the role of Sla2 / HIP family in endocytosis was proposed [6,14], it is possible that rHIP1R may also be involved in endocytosis of certain cell lineages that express this protein. The I / LWEQ module is known to act as an F-actinbinding module and connect with diverse proteins involved in distinct cellular processes, such as cell adhesion, cytoskeletal organization and cell differentiation, to the actin cytoskeleton [19]. Indeed, Sla2 and its homologs are expected to play a role in the organization of some plasma membrane proteins by linking them to the actin cytoskeleton through the I / LWEQ module [18]. Thus, rHIP1R may play a role for linking the endocytotic machinery to the actin cytoskeleton [6]. Although HIP1 and HIP1R are structurally related, they may have distinct roles in vivo. Firstly, HIP1R does not interact with huntingtin as examined by two-hybrid assays [4], whereas HIP1 does [13]. The lack of HIP1R interaction with huntingtin was supported by our study examining co-immunoprecipitation (not shown) and by the finding that HIP1Rs do not show homology to the huntingtinbinding domain (Fig. 1c). Secondly, phenylalanine-398 in HIP1 DED, a critical amino acid for apoptosis-inducing activity, was replaced by glutamate in the corresponding region of rat, mouse and human HIP1Rs. This suggests that rHIP1R may not have any apoptosis-inducing activity. Thus, rHIP1R, differently from HIP1, may be a nonproapoptotic member of HIPs [4]. These findings suggest a differential role of HIP1 and HIP1R towards huntingtinrelated neuronal events. It is possible that HIP1R protein is involved in the huntingtin function, although it does not appear to interact with huntingtin directly, because HIP1R can form a dimer or oligomer with huntingtin-interacting HIP protein [4]. If HIP1R interacts competitively with HIP1, as suggested by Chopra et al. [4], the binding of HIP1 homo-oligomer to huntingtin may be reduced by the production of HIP1HIP1R hetero-oligomer. Thus, HIP1R could be a potential modulator of HIP activity, affecting on HIP1-inducing apoptosis [4] and / or membrane trafficking that involve HIP1 and huntingtin.

4.1. Structural analyses of rHIP1 R 4.2. Postsynaptic role of rHIP1 R We isolated from a rat brain cDNA library a rHIP1R cDNA clone, which encodes a Sla2 / HIP family protein. The deduced rHIP1R protein had ENTH and I / LWEQ domains in the N-terminus and the C-terminus, respectively, as do other Sla2 / HIP family proteins. Sla2 / HIP1 family proteins share the N-terminus with epsin and Cterminus with talin. The C-terminal portion of rHIP1R interacted with F-actin (Fig. 5) as expected by the presence of an F-actin-binding module, I / LWEQ module [18]. The amino acid sequence of HIP1R was highly conserved in rats, mice and humans, and major structural domains were well conserved with other Sla2 / HIP family members.

We showed in the present study that the rHIP1R protein was localized in the postsynaptic structure at the electron microscopic level. Interestingly, huntingtin is also localized at synaptic sites as shown by electron microscopic examination [5,9]. Distribution of huntingtin in the postsynaptic structures (Fig. 5h in Ref. [5]) is quite similar to that of rHIP1R at the electron microscopic level. The postsynaptic localization of huntingtin is supported by the finding that normal huntingtin binds to PSD-95 [30]. HIP1 is enriched in the brain and the subcellular distribution of this protein was shown to be nearly identical to that of

222

A. Okano et al. / Brain Research 967 (2003) 210–225

Fig. 9. Electron microscopic examination of rHIP1R in the rat cerebral cortex (electron-stained with lead citrate and uranium). Den, p (either upper or lower case letters), asterisks, arrows, and open arrows indicate dendrite, PSD, presynaptic terminals, immunoreactive areas apart form PSD, and small vesicular structures stained, respectively. Immunohistochemistry using preabsorbed antibody is shown in (a). Pictures are at the same magnification. Bar in (f), 200 nm.

A. Okano et al. / Brain Research 967 (2003) 210–225

223

concomitant gene expression of rHIP1R and huntingtin during rat development (Fig. 2b). Thus, it is possible that huntingtin, its binding protein HIP1 and HIP1R are colocalized in vivo, at least at the synapse, and work together at the synapse. The present electron microscopic findings showed an association of the rHIP1R immunoreactivity with PSD, whereas the protein was not detected in the PSD fraction by Western blotting. The rHIP1R may be dissociated from PSD by Triton X-100 treatment and subsequent sucrose gradient centrifugation. This may be due to the weak association of the protein with PSD or indirect interactions such as those through SPM or intracellular membranous organelles. We found that rHIP1R-immunoreactivities are localized to the vesicular structures, either very close to or separate from PSD in the spines and dendrites (Figs. 9 and 10). These immunoreactive vesicular structures are fragments of the endoplasmic reticulum (ERs) or parts of endosomes in the postsynaptic regions, which were suggested to be the sites for vesicular transport of postsynaptic proteins [27]. These structures might be a part of the clathrin-coated vesicle pathways, since it is well established that HIP1R / Hip12 and HIP1 are endocytotic proteins associated with clathrin-coated vesicles [7,20,21,34]. The clathrin coated vesicle system occurs in the postsynaptic side of the synapse in addition to the presynaptic side [8]. Huntingtin is suggested to have a role for the vesicle trafficking within neuronal cells [5,33]. These findings suggest that the HIP1R and HIP1, together with huntingtin, play a role for endocytotic vesicular trafficking at the postsynaptic side of the synapse.

Acknowledgements This research was supported in part by a Grant-in-Aid for Scientific Research from the Japanese Ministry of Education, Science and Culture, Toyota Physical and Chemical Research Institute, Uehara Memorial Foundation, and Joint Research Project under The Japan–Korea Basic Scientific Cooperation Program of Japan Society for the Promotion of Science.

References Fig. 10. Electron microscopic examination of rHIP1R in the rat cerebral cortex (not stained with lead citrate and uranium). Asterisks, triangles, open arrowhead, an arrow, and open arrows indicate presynaptic terminals, PSD stained faintly or moderately, unstained PSD, immunoreactive areas apart form PSD, and small vesicular structures stained, respectively. Pictures are at the same magnification. Bar in (a), 200 nm.

huntingtin [13,35]. We also found that rHIP1R and huntingtin share quite similar distribution patterns in synaptic fractions of rat brain (Fig. 7). Moreover, we found

[1] P.G. Bhide, M. Day, E. Sapp, C. Schwarz, A. Sheth, J. Kim, A.B. Young, J. Penney, J. Golden, N. Aronin, M. DiFiglia, Expression of normal and mutant huntingtin in the developing brain, J. Neurosci. 16 (1996) 5523–5535. [2] J.R. Burke, J.J. Enghild, M.E. Martin, Y.S. Jou, R.M. Myers, A. D Roses, J.M. Vance, W.J. Strittmatter, Huntingtin and DRPLA proteins selectively interact with the enzyme GAPDH, Nat. Med. 2 (1996) 347–350. [3] A.H. Chishti, A.C. Kim, S.M. Marfatia, M. Lutchman, M. Hanspal, H. Jindal, S.C. Liu, P.S. Low, G.A. Rouleau, N. Mohandas, J.A. Chasis, J.G. Conboy, P. Gascard, Y. Takakuwa, S.C. Huang, E.J.

224

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17] [18]

A. Okano et al. / Brain Research 967 (2003) 210–225 Benz Jr., A. Bretscher, R.G. Fehon, J.F. Gusella, V. Ramesh, F. Solomon, V.T. Marchesi, S. Tsukita, S. Tsukita, M. Arpin, D. Louvard, N.K. Tonks, J.M. Anderson, A.S. Fanning, P.J. Bryant, D.F. Woods, K.B. Hoover, The FERM domain, a unique module involved in the linkage of cytoplasmic proteins to the membrane, Trends Biochem. Sci. 23 (1996) 281–282. V.S. Chopra, M. Metzler, D.M. Raspe, A.E. Engqvist-Goldstein, R. Singaraja, L. Gan, K.M. Fichter, K. McCutcheon, D. Drubin, D.W. Nicholson, M.R. Hayden, HIP12 is a non-proapoptotic member of a gene family including HIP1, an interacting protein with huntingtin, Mamm. Genome 11 (2000) 1006–1015. M. DiFiglia, E. Sapp, K. Chase, C. Schwarz, A. Meloni, C. Young, E. Martin, J.P. Vonsattel, R. Carraway, S.A. Reeves, F.M. Boyce, N. Aronin, Huntingtin is a cytoplasmic protein associated with vesicles in human and rat brain neurons, Neuron 14 (1995) 1075–1081. A.E. Engqvist-Goldstein, M.M. Kessels, V.S. Chopra, M.R. Hayden, D.G. Drubin, An actin-binding protein of the Sla2 / Huntingtin interacting protein 1 family is a novel component of clathrin-coated pits and vesicles, J. Cell Biol. 147 (1999) 1503–1518. A.E. Engqvist-Goldstein, R.A. Warren, M.M. Kessels, J.H. Keen, J. Heuser, D.G. Drubin, The actin-binding protein Hip1R associates with clathrin during early stages of endocytosis and promotes clathrin assembly in vitro, J. Cell Biol. 154 (2001) 1209–1223. N. Eshhar, R.S. Petralia, C.A. Winters, A.S. Niedzielski, R.J. Wenthold, The segregation and expression of glutamate receptor subunits in cultured hippocampal neurons, Neuroscience 57 (1993) 943–964. C.A. Gutekunst, A.I. Levey, C.J. Heilman, W.L. Whaley, H. Yi, N.R. Nash, H.D. Rees, J.J. Madden, S.M. Hersch, Identification and localization of huntingtin in brain and human lymphoblastoid cell lines with anti-fusion protein antibodies, Proc. Natl. Acad. Sci. USA 92 (1995) 8710–8714. A.S. Hackam, A.S. Yassa, R. Singaraja, M. Metzler, C.-A. Gutekunst, L. Gan, S. Warby, C.L. Wellington, J. Vaillancourt, N. Chen, F.G. Gervais, L. Raymond, D.W. Nicholson, M.R. Hayden, Huntingtin interacting protein 1 induces apoptosis via a novel caspase-dependent death effector domain, J. Biol. Chem. 275 (2000) 41299–41308. D.A. Holtzman, S. Yang, D.G. Drubin, Synthetic-lethal interactions identify two novel genes, SLA1 and SLA2, that control membrane cytoskeleton assembly in Saccharomyces cerevisiae, J. Cell Biol. 122 (1993) 635–644. M.A. Kalchman, R.K. Graham, G. Xia, H.B. Koide, J.G. Hodgson, K.C. Graham, Y.P. Goldberg, R.D. Gietz, C.M. Pickart, M.R. Hayden, Huntingtin is ubiquitinated and interacts with a specific ubiquitin-conjugating enzyme, J. Biol. Chem. 271 (1996) 19385– 19394. M.A. Kalchman, H.B. Koide, K. McCutcheon, R.K. Graham, K. Nichol, K. Nishiyama, P. Kazemi-Esfarjani, F.C. Lynn, C. Wellington, M. Metzler, Y.P. Goldberg, I. Kanazawa, R.D. Gietz, M.R. Hayden, HIP1, a human homologue of S. cerevisiae Sla2p, interacts with membrane-associated huntingtin in the brain, Nat. Genet. 16 (1997) 44–53. B.K. Kay, M. Yamabhai, B. Wendland, S.D. Emr, Identification of a novel domain shared by putative components of the endocytic and cytoskeletal machinery, Protein Sci. 8 (1999) 435–438. X.J. Li, S.H. Li, A.H. Sharp, F.C. Nucifora Jr., G. Schilling, A. Lanahan, P. Worley, S.H. Snyder, C.A. Ross, A huntingtin-associated protein enriched in brain with implications for pathology, Nature 378 (1995) 398–402. W. Li, A. Okano, Q.B. Tian, K. Nakayama, T. Furihata, H. Nawa, T. Suzuki, Characterization of a novel synGAP isoform, synGAP-b, J. Biol. Chem. 276 (2001) 21417–21424. A. Lupas, Prediction and analysis of coiled-coil structures, Methods Enzymol. 266 (1996) 513–525. R.O. McCann, S.W. Craig, The I / LWEQ module, a conserved sequence that signifies F-actin binding in functionally diverse

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28] [29]

[30]

[31]

[32]

[33]

[34]

[35]

proteins from yeast to mammals, Proc. Natl. Acad. Sci. USA 94 (1997) 5679–5684. R.O. McCann, S.W. Craig, Functional genomic analysis reveals the utility of the I / LWEQ module as a predictor of protein:actin interaction, Biochem. Biophys. Res. Commun. 266 (1999) 135–140. M. Metzler, V. Legendre-Guillemin, L. Gan, V. Chopra, A. Kwok, P.S. McPherson, M.R. Hayden, HIP1 functions in clathrin-mediated endocytosis through binding to clathrin and adaptor protein 2, J. Biol. Chem. 276 (2001) 39271–39276. S.K. Mishra, N.R. Agostinelli, T.J. Brett, I. Mizukai, T.S. Ross, L.M. Traub, Clathrin- and AP-2-binding sites in HIP1 uncover a general assembly role for endocytic accessory proteins, J. Biol. Chem. 276 (2001) 46230–46236. S. Murata, N. Usuda, A. Okano, S. Kobayashi, T. Suzuki, Occurrence of a transcription factor, signal transducer and activators of transcription 3 (Stat3), in the postsynaptic density of the rat brain, Mol. Brain Res. 78 (2000) 80–90. T. Nakamura, K. Takeuchi, S. Muraoka, H. Takezoe, N. Takahashi, N. Mori, A neurally enriched coronin-like protein. clipinC, is a novel candidate for an actin cytoskeleton-cortical membrane-linking protein, J. Biol. Chem. 274 (1999) 13322–13327. J. Schultz, R.R. Copley, T. Doerks, C.P. Ponting, P. Bork, SMART: a Web-based tool for the study of genetically mobile domains, Nucleic Acids Res. 28 (2000) 231–234. N. Seki, M. Muramatsu, S. Sugano, Y. Suzuki, A. Nakagawara, M. Ohhira, A. Hayashi, T. Hori, T. Saito, Cloning, expression analysis, and chromosomal localization of HIP1R, an isolog of huntingtin interacting protein (HIP1), J. Hum. Genet. 43 (1998) 268–271. I. Schmitt, D. Bachner, D. Megow, P. Henklein, H. Hameister, J.T. Epplen, O. Riess, Expression of the Huntington disease gene in rodents: cloning the rat homologue and evidence for downregulation in nonneuronal tissues during development, Hum. Mol. Genet. 4 (1995) 1173–1182. J. Spacek, K.M. Harris, Three-dimensional organization of smooth endoplasmic reticulum in hippocampal CA1 dendrites and dendritic spines of the immature and mature rat, J. Neurosci. 17 (1997) 190–203. J.A. Spudich, S. Watt, The regulation of rabbit skeletal muscle contraction, J. Biol. Chem. 246 (1971) 4866–4871. P.C. Suen, K. Wu, J.L. Xu, S.Y. Lin, E.S. Levine, I.B. Black, NMDA receptor subunits in the postsynaptic density of rat brain: expression and phosphorylation by endogenous protein kinases, Mol. Brain Res. 59 (1998) 215–228. Y. Sun, A. Savanenin, P.H. Reddy, Y.F. Liu, Polyglutamine-expanded Huntingtin promotes sensitization of N-methyl-D-aspartate receptors via post-synaptic density 95, J. Biol. Chem. 276 (2001) 24713–24718. T. Suzuki, N. Usuda, H. Ishiguro, S. Mitake, T. Nagatsu, K. Okumura-Noji, Occurrence of a transcription factor, cAMP response element-binding protein (CREB), in the postsynaptic sites of the brain, Mol. Brain Res. 61 (1998) 69–77. T. Suzuki, N. Usuda, S. Murata, A. Nakazawa, K. Ohtsuka, H. Takagi, Presence of molecular chaperones, heat shock cognate (Hsc) 70 and heat shock protein (Hsp) 40, in the postsynaptic structures of rat brain, Brain Res. 816 (1999) 99–110. T. Tukamoto, N. Nukina, K. Ide, I. Kanazawa, Huntington’s disease gene product, huntingtin, associates with microtubules in vitro, Mol. Brain Res. 51 (1997) 8–14. S. Waelter, E. Scherzinger, R. Hasenbank, E. Nordhoff, R. Lurz, H. Goehler, C. Gauss, K. Sathasivam, G.P. Bates, H. Lehrach, E.E. Wanker, The huntingtin interacting protein HIP1 is a clathrin and a-adaptin-binding protein involved in receptor-mediated endocytosis, Hum. Mol. Genet. 10 (2001) 1807–1817. E.E. Wanker, C. Rovira, E. Scherzinger, R. Hasenbank, S. Walter, D. Tait, J. Colicelli, H. Lehrach, HIP-I: a huntingtin interacting protein isolated by the yeast two-hybrid system, Hum. Mol. Genet. 6 (1997) 487–495.

A. Okano et al. / Brain Research 967 (2003) 210–225 [36] R. Wilson, R. Ainscough, K. Anderson, C. Baynes, M. Berks, J. Bonfield, J. Burton, M. Connell, T. Copsey, J. Cooper, A. Coulson, M. Craxton, S. Dear, Z. Du, R. Durbin, A. Favello, A. Fraser, L. Fulton, A. Gardner, P. Green, T. Hawkins, L. Hillier, M. Jier, L. Johnston, M. Jones, J. Kershaw, J. Kirsten, N. Laisster, P. Latreille, J. Lightning, C. Lloyd, B. Mortimore, M. O’callaghan, J. Parsons, C. Percy, L. Rifken, A. Roopra, D. Saunders, R. Shownkeen, M. Sims, N. Smaldon, A. Smith, M. Smith, E. Sonnhammer, R. Staden, J.

225

Sulston, J. Thierry-Mieg, K. Thomas, M. Vaudin, K. Vaughan, R. Waterson, A. Watson, L. Weinstock, J. Wilkinson-Sproat, P. Wohldman, 2.2 Mb of contiguous nucleotide sequence from chromosome III of C. elegans, Nature 368 (1994) 32–38. [37] J.D. Wood, J.C. Macmillan, P.S. Harper, Huntingtin and apparent variations in the subcellular localisation of huntingtin in human, mouse and rat brain, Hum. Mol. Genet. 5 (1996) 481–487.