Characterization of the 105-kDa molecular chaperone Identification, biochemical properties, and localization Mika Matsumori 1,2 , Hideaki Itoh 1 , Itaru Toyoshima 3 , Atsushi Komatsuda 4 , Ken-ichi Sawada 4 , Jun Fukuda 5 , Toshinobu Tanaka 5 , Atsuya Okubo 6 , Hiroyuki Kinouchi 6 , Kazuo Mizoi 6 , Tokiko Hama 7 , Akira Suzuki 1 , Fumio Hamada 8 , Michiro Otaka 3 , Yutaka Shoji 2 and Goro Takada 2 1 Department of Biochemistry, 2 Department of Pediatrics, 3 First Department of Internal Medicine, 4 Third Department of Internal Medicine, 5 Department of Gynecology, and 6 Department of Neurosurgery, Akita University School of Medicine, Akita City, Japan; 7 President’s Frontier Laboratory. Mitsubishi Kasei Institute of Life Sciences, Tokyo, Japan; 8 Department of Material-Process Engineering and Applied Chemistry for Environment, Akita University Faculty of Engineering and Resource Science, Akita City, Japan We have characterized the biochemical properties of the testis and brain-specific 105-kDa protein which is cross- reacted with an anti-bovine HSP90 antibody. The protein was induced in germ cells by heat stress, resulting in a protein which is one of the heat shock proteins [Kumagai, J., Fuk- uda, J., Kodama, H., Murata, M., Kawamura, K., Itoh, H. & Tanaka, T. (2000) Eur. J. Biochem. 267, 3073–3078]. In the present study, we characterized the biochemical properties of the protein. The 105-kDa protein inhibited the aggregation of citrate synthase as a molecular chaperone in vitro.ATP/ MgCl 2 has a slight influence of the suppression of the citrate synthase aggregation by the 105-kDa protein. The protein possessed chaperone activity. The protein was able to bind to ATP–Sepharose like the other molecular chaperone HSP70. A partial amino-acid sequence (24 amino-acid residues) of the protein was determined and coincided with those of the mouse testis- and brain-specific APG-1 and osmotic stress protein 94 (OSP94). The 105-kDa protein was detected only in the medulla of the rat kidney sections similar to OSP94 upon immunoblotting. The purified 105-kDa protein was cross-reacted with an antibody against APG-1. These results suggested that APG-1 and OSP94 are both identical to the 105-kDa protein. There were highly homologous regions between the 105-kDa protein/APG-1/OSP94 and HSP90. The region of HSP90 was also an immunoreactive site. An anti-bovine HSP90 antibody may cross-react with the 105-kDa protein similar to HSP90 in the rat testis and brain. We have investigated the localization and developmental induction of the protein in the rat brain. In the immuno- histochemical analysis, the protein was mainly detected in the cytoplasm of the nerve and glial cells of the rat brain. Although the 105-kDa protein was localized in all rat brain segments, the expression pattern was fast in the cerebral cortex and hippocampus and slow in the cerebellum. Keywords: molecular chaperone; 105-kDa protein; APG-1; OSP94. All living cells display a rapid induction of some proteins known as molecular chaperones (heat-shock proteins or stress proteins) when the cells are exposed to environmental stresses such as lethal heat shock or variety of toxic reagents [1]. Among the molecular chaperones, HSP90 is a cytoplas- mic protein in unstressed mammalian cells and has been found in transient association with steroid hormone recep- tors and regulates their activation mechanism [2]. It has been reported that HSP90 plays the role of capacitor for morphological evolution [3] and facilitates the synthesis and correct folding of other intracellular proteins [4]. HSP90 contains two independent chaperone sites both in the N-terminal and C-terminal of the protein [5,6]. Each chaperone activity of the protein will be inhibited by different antineoplastic agents of geldanamycin and cisplatin, respectively [7]. We reported before that a rat 105-kDa testis protein was cross-reacted with an antibody against bovine HSP90 [8]. The 105-kDa protein was also detected in the brain, but not in the liver, lung, spleen, kidney, ovarium, or uterus, in contrast to the wide distribution of HSP90. There was a high similarity between HSP90 and the 105-kDa protein on peptide mapping with trypsin digestion. Except for the molecular mass, the physicochemical properties of the 105-kDa protein were similar to those of HSP90, and theproteinseemstobeacognateproteinofHSP90. On immunoblotting analysis, the 105-kDa protein appeared at approximately the age of 5 weeks and coincided with the appearance of spermatozoa during the development of the rat testis. The 105-kDa protein was more abundant in the spermatozoa but not in a somatic cell line derived from a Correspondence to H. Itoh, Department of Biochemistry, Akita University School of Medicine, 1-1-1 Hondo, Akita City, 010-8543 Japan. Tel.: + 81 18 884 6078, Fax: + 81 18 884 6078, E-mail: hideaki@med.akita-u.ac.jp Abbreviations: HSP110, 105, 90, HSP70 and HSP60, 110-, 105-, 90-, 70-, and 60-kDa heat shock proteins; GRP78, 78-kDa glucose regulated protein; OSP94, osmotic stress protein 94; HSF-1, heat shock transcription factor 1; CS, citrate synthase; AzC, L -azetidine- 2-carboxylic acid; PC12, rat phenochromocytoma cell; BCIP, 5-bromo-4-chloro-3-indolyphosphate p-toluidine salt; NBT, nitroblue tetrazolium chloride. (Received 29 July 2002, revised 13 September 2002, accepted 18 September 2002) Eur. J. Biochem. 1–10 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03272.x Leydig cell tumor in rat testis [9]. These results indicate that the 105-kDa protein is one of the sperm-specific proteins. Recently, we have reported that signals of the 105-kDa protein were selectively detected immunohistochemically in the germ cells and might translocate into the nuclei from the cytoplasm in response to heat shock [10]. Moreover, the 105-kDa protein formed a complex with p53 at 32.5 °C, which is the scrotal temperature, but not at 37 °C, which is the suprascrotal temperature; the 105-kDa protein is sug- gested to contribute to the regulation of p53 function in testicular germ cells [10]. It has been shown that testis-specific APG-1 (accession no. D49482) and osmotic stress protein 94 (OSP94) (accession no. U23291) cDNA were cloned independently [11,12]. Both APG-1 and OSP94 are members of the HSP110/SSE subfamily. The cDNA of both APG-1 and OSP94 encodes 838 amino-acid residues, and those amino- acid sequences were the same. To determine the interaction between the 105-kDa protein and APG-1 or OSP94, we investigated the amino-acid sequence and some biochemical properties of the 105-kDa protein. In the present study, we discuss the biochemical properties of the 105-kDa protein in the rat brain and the interaction between the 105-kDa protein and APG-1 or OSP94. In our earlier studies, the 105-kDa protein was induced in germ cells by heat stress, and the protein formed a complex with p53 in a temperature-dependent manner [10]. The protein appeared at 5 weeks (postbirth), during the devel- opment of the rat testis and coincided with the appearance of spermatozoa [9]. The 105-kDa protein may be involved in the regeneration of p53 function in testicular germ cells. In contrast, the localization of the protein in the brain has not yet been known. We have also investigated the distribution of the 105-kDa protein in the rat brain and the appearance of 105-kDa protein in several sections of the post natal developmental rat brain. MATERIALS AND METHODS ATP–Sepharose was prepared as described previously [13]. DE-52 was obtained from Whatman and Lysyl endopep- tidase (EC 3.4.21.50) was from Wako Pure Chemical Institute. Purification of the 105-kDa protein The 105-kDa testis and brain protein was purified from rat testis as described previously [8]. The protocols for animal experimentation described in this paper were previously approved by the Animal Research Committee, Akita University School of Medicine; the ÔGuidelines for Animal ExperimentationÕ of the University were completely adhered to in all subsequent animal experiments. Antibody production An antibody against to the rat 105-kDa protein was produced by intramuscular injection into a rabbit of 1 mg of the protein emulsified in complete Freund’s adjuvant. Booster shots were given three times in the same manner as the original injection at 2-week intervals. The rabbit was bled 10 days after the last injection. The antiserum raised against rat 105-kDa protein (2 mL) was dialyzed against 10 m M Tris/HCl (pH 7.4). The serum was applied to a DEAE–cellulose column (1 · 4 cm) pre-equilibrated in 10 m M Tris/HCl (pH 7.4). The pass-through fractions were collected and examined as IgG on SDS/PAGE gel electrophoresis. An anti-(rat 105-kDa protein) IgG was used in this study. Antibodies against porcine HSP60, bovine HSP70, and bovine HSP90 were used as described previ- ously [8,13,14]. Measurement of in vitro chaperone activity The influence of the 105-kDa protein on the thermal aggregation of mitochondrial citrate synthase (CS; EC 4.1.37; Boehringer–Mannheim) at 43 °C was measured as previously described [7]. Light scattering CS (0.075 l M ) in 50 m M Hepes buffer (pH 7.4) in the presence or absence of bovine serum albumin (15 l M ) and the 105-kDa protein (0.075 l M ) was monitored for 90 min by the optical density at 500 nm in a Pharmacia Ultospec 3000 UV/Vis spectro- photometer equipped with a temperature control unit using semi microcuvettes (1 mL) with a path length of 10 mm. ATP–Sepharose column chromatography Rat testes were homogenized in 9 vols of 50 m M Tris/HCl (pH 7.4) containing 0.25 M sucrose and centrifuged at 9000 g for 10 min at 4 °C. The supernatant was collected, followed by centrifugation at 105 000 g for 60 min at 4 °C. The supernatant obtained from ultracentrifugation was used as the rat testis cytosol in the present study. ATP–Sepharose was equilibrated in buffer A (10 m M Tris/HCl, pH 7.4, 5 m M CaCl 2 ,5m M MgCl 2 ). Rat testis cytosols containing 5 m M CaCl 2 and 5 m M MgCl 2 were applied to the column and washed with buffer A containing 0.15 M NaCl. After washing the column, binding proteins were eluted with 5 m M ATP in buffer A. Eluted proteins were detected by SDS/PAGE and immunoblotting. Amino-acid sequence of 105-kDa protein The amino-acid sequence of 105-kDa protein was deter- mined using a protein sequencer as described previously [14,15]. Briefly, purified 105-kDa protein from the testis was electrophoresed by SDS/PAGE, and the protein band was excised and digested using lysyl endopeptidase. The reverse phase column (Wakosil 5C 18 , Wakopak) that was connected to an HPLC apparatus purified the digested peptides. The peptides were applied onto the column and eluted with a linear gradient of 0–60% acetonitrile (v/v) in 0.1% trifluoroacetic acid (v/v) at a flow rate of 0.5 mLÆmin )1 , and 0.5-mL fractions were collected. Amino-acid sequencing of the purified peptides was performed with a 491 Procise protein sequence system (Perkin-Elmer). Gel electrophoresis SDS/PAGE was carried out by the procedure of Laemmli [16]. Gels were stained with 0.1% Coomassie Brilliant Blue (v/v) with 25% isopropyl alcohol (v/v) and 10% acetic acid (v/v) and destained with 10% isopropyl alcohol (v/v) and 10% acetic acid (v/v). 2 M. Matsumori et al. (Eur. J. Biochem.) Ó FEBS 2002 Immnoblotting Samples were electrophoresed on SDS/polyacrylamide gels, transferred to a poly(vinylidene difluoride) membrane (Bio-Rad) electrophoretically, and processed as described by Towbin et al. [17]. The membrane was incubated with anti-HSP90 antibody (diluted 1 : 1000 in 7% skim milk) or anti-rat 105-kDa protein IgG (diluted 1 : 2000 in 7% skim milk). The membrane was also incubated with an antibody against APG-1 or an antibody against HSF-1 (Santa Cruz Biotechnology, diluted 1 : 500 in 7% skim milk). The membranes were treated with alkaline phosphatase anti- (rabbit IgG) Ig (Bio-Rad) (diluted 1 : 1000 in 7% skim milk). Antigen-antibody complexes were visualized by reacting alkaline phosphatase using 5-bromo-4-chloro- 3-indolyphosphate p-toluidine salt (BCIP) and nitroblue tetrazolium chloride (NBT). Water-restriction of rat Control Wister rats (Male, 6 weeks) were provided ad libitum access to water, and dehydrated rats were restricted from drinking water for 3 and 5 days. All rats were sacri- ficed, the kidneys were removed, and the renal cortex, medulla, and papilla were dissected. Tissues were, respect- ively, homogenized in 3 vols of 10 m M Tris/HCl (pH 7.4) containing 0.15 M NaCl and centrifuged at 20 000 g for 10 min at 4 °C. The supernatants were used for SDS/PAGE and immunoblotting as the soluble fraction. The precipi- tates were washed with the same buffer, collected by centrifugation at 20 000 g for 10 min at 4 °C, and used for SDS/PAGE and immunoblotting as the insoluble fraction. Homology search, hydropathy profile, and secondary structure prediction DNASIS (version 2.1, Hitachi Software Engineering Co, Ltd) was used for the amino-acid sequence homology search, hydropathy profiles, and secondary structure prediction of the 105-kDa protein, HSP90, APG-1, and OSP94. For the hydropathy profiles of APG-1 or OSP94, we used the hydrotable of Hopp and Woods. The secondary structure prediction of the proteins has been performed by Chou and Fasman prediction methods. Detection of the 105-kDa protein in rat brain during the development Female and male Wistar rats were born from the same parents, 3.5 days old, 1-, 2-, 3-, 4-, 5-, and 6-weeks-old, and dissected to obtain several brain sections. The sections were the olfactory lobe, cerebral cortex, hippocampus, mid brain, cerebellum, and medulla oblongata. Each tissue section was homogenized in 3 vols of 10 m M Tris/ HCl (pH 7.4) containing 0.15 M NaCl and centrifuged at 40 000 g for 10 min at 4 °C. The supernatants were used for SDS/PAGE and immunoblotting as the soluble fraction. The precipitates were washed with the same buffer, collected by centrifugation at 40 000 g for 10 min at 4 °C, and used for SDS/PAGE and immunoblotting as the insoluble fraction. Control Wister rat (Female, 6 weeks) were sacrificed and the brain was removed. The olfactory lobe, cortex of the cerebrum, hippocampus, mid brain, cerebellum, and medulla oblongata were dissected. Tissues were, respectively, homogenized as described above. The supernatant and precipitates were used for SDS/PAGE or immunoblotting. Immunohistochemistry Nonfixed and quickly frozen rat brains were sectioned in a cryostat. Sections (15 lm thick) were fixed 3.7% formalde- hyde in NaCl/P i for 10 min at room temperature. After the sections were washed with NaCl/P i (5 min, 3 times), they were incubated with anti-rat 105-kDa protein IgG (diluted 1 : 200 in 0.2% BSA/0.01% Saponin NaCl/P i )overnightat 4 °C and then washed again with NaCl/P i (15 min, threefold). Bound antibodies were visualized with horse- radish peroxidase-labeled anti-(rabbit IgG) Ig conjugated with amino-acid polymer (Histofine simple stain MAX-PO, Nichirei) according to the manufacturers instructions. Cell culture PC12 and PC12h cells (provided from T. Hama, Mitsubishi Kasei Institute of Life Sciences) were grown in RPMI-1640 supplemented with 10% fetal calf serum in a culture flask at 37 °C in a humidified atmosphere containing 5% CO 2 . Upon approaching confluence (c.1· 10 6 cells/culture flask), the medium was removed, and the cells were stressed by the addition of 5 m ML -azetidine 2-carboxylic acid (AzC) in fresh medium containing 10% fetal bovine serum for 6 h. The cells were stressed by increasing the temperature to 43 °Cfor30minfollowedbygrowthat37°Cfor6h.The cells were harvested and washed twice with 10 m M Tris/ HCl, pH 7.4) containing 0.125 M NaCl. After the centrif- ugation at 2000 g for 10 min, the cells were collected. The pellet was then sonicated in SDS sample buffer and centrifuged at 18 000 g for 10 min. The supernatant was used for SDS/PAGE and immunoblotting. RESULTS Specificity of antibody against rat 105-kDa protein Two types of sample were prepared as the soluble and insoluble fractions from the cortex of the cerebrum. Samples were separated on SDS/PAGE following to immunoblotting using anti-bovine HSP90 antibody [8] or anti-rat 105-kDa protein IgG (Fig. 1). An antibody against bovine HSP90 was recognized as rat HSP90 (Fig. 1B). HSP90 was detected mainly in the soluble fraction of the rat brain. The antibody recognized another protein with a molecular mass of 105 kDa. Anti-bovine HSP90 reacts mainly with HSP90 and also reacts faintly with 105-kDa protein. On the contrary, anti-rat 105-kDa protein IgG was cross-reacted with only 105-kDa protein (Fig. 1C). No other protein bands were detected both in the soluble and insoluble fractions of the rat brain. The antibody is highly specific for the antigen. Thus, an anti-rat 105-kDa protein IgG can strongly recognize the protein compared to an anti-bovine HSP90 antibody. Chaperone activity of the 105-kDa protein To analyze the functional properties of the 105-kDa protein, we studied its action in protein folding in vivo. As shown Ó FEBS 2002 Biochemical properties of the 105-kDa protein (Eur. J. Biochem.)3 Fig. 2, spontaneous aggregation of CS has been obserbed at 43 °C. Although there was no effect on the reaction in the presence of 200-fold molar excess bovine serum albumin, an equimolar amount of the 105-kDa protein suppressed the aggregation of CS. Next, we investigated the influence of ATP on the assay system. ATP/MgCl 2 has a slight influence on the suppression of CS aggregation by the 105-kDa protein. The 105-kDa protein apparently interacts transiently with the highly structured early unfolding inter- mediates. ATP–Sepharose column chromatography Among the mammalian molecular chaperones, HSP70, GRP78, and HSP60 are able to bind to an ATP–Sepharose column [13,14]. Although HSP90 is not able to bind to ATP–Sepharose, the protein has two independent ATP- binding sites in both the N- and C-terminals [18]. As mentioned above, almost all molecular chaperones can interact with ATP. On the contrary, there are no reports for ATP-binding of 105-kDa protein. To investigate interaction between the 105-kDa protein and ATP, we analyzed the ATP-binding proteins of rat using ATP–Sepharose column chromatography. Rat testis cytosols were applied onto the column; eluted proteins were detected by SDS/PAGE and immunoblotting as described in Materials and methods. As shown in Fig. 3A, we detected some protein bands with molecular masses of 70-, 78-, and 105-kDa on the gel. On immunoblotting using anti-HSP90 antibody, the 105-kDa protein was detected only in the one eluted fraction to some extent (Fig. 3B). HSP90 was also detected in the same fraction with a very faint protein band. On the other hand, the 105-kDa protein was detected in all eluted fractions on immunoblotting using an anti-rat 105-kDa protein IgG (Fig. 3C). Because of a high titer of an anti-(rat 105-kDa protein) Ig, IgG was recognized in all eluants in spite of the low concentration on immunoblotting. Due to the low Fig. 2. Chaperone activity of the 105-kDa protein. Thermal aggrega- tion of CS (0.075 l M ) in the absence of additional components (open circle), in the presence of a 200-fold molar ratio of albumin (closed circle), an equimolar ratio of the 105-kDa protein (open triangle), an equimolar ratio of the 105-kDa protein and 5 m M ATP/MgCl 2 (closed triangle). ATP/MgCl 2 was added (down arrow) after 30 min of ther- mal aggregation of CS and the 105-kDa protein (open square). Fig. 3. ATP–Sepharose column chromatography of testis cytosols. The eluted fractions from an ATP–Sepharose column were electrophoresed on SDS-polyacrylamide gels (9% gel), which were stained with Coo- massie Brilliant Blue (A), by immunoblotting with an antibody against bovine HSP90 (B), or by immunoblotting with an antibody against the 105-kDa protein IgG (C). Fig. 1. Specificity of an antibody against the 105-kDa protein. Soluble and insoluble fraction of rat brain were electrophoresed on SDS- polyacrylamide gels (10% gel), which were stained with Coomassie Brilliant Blue (A), by immunoblotting with an antibody against bovine HSP90 (B), or by immunoblotting with an antibody against the 105-kDa protein IgG (C). 4 M. Matsumori et al. (Eur. J. Biochem.) Ó FEBS 2002 concentration, the protein band was faintly recognized only in one fraction on the Coomassie Brilliant Blue stained gel. An anti-HSP90 antibody could barely recognize the protein in only one eluted fraction. These results suggested that the 105-kDa protein is an ATP-binding protein the same as the other molecular chaperones like a HSP70 and GRP78. Amino-acid sequence of the 105-kDa protein To investigate the biochemical properties of the 105-kDa protein, the amino-acid sequence of the protein was determined using a peptide sequencer. The peptides of the 105-kDa protein digested with lysyl endopeptidase were purified using the reverse phase column connected to an HPLC. As shown in Fig. 4, the two peptides (#40 and #70) were sequenced. Nine amino-acid residues were obtained from peptide #40, and 15 amino-acid residues were obtained from peptide #70. The total of 24 amino-acid sequences from the two peptides of the protein had complete similarity to APG-1, a testis-specific protein [11], and OSP94, a renal medulla-specific protein [12]. No homology has been shown between the 105-kDa protein and HSP105 [19]. The amino- acid sequence of the 105-kDa protein showed that the protein is identical to APG-1 and OSP94. Renal localization of the 105-kDa protein during water restriction It has been shown that OSP94 mRNA is induced in the renal inner medulla of a water-restricted mouse [12]. We could detect the 105-kDa protein only in the brain and testis on immunoblotting. Although a partial amino-acid sequence of the 105-kDa protein coincides with that of OSP94, we could not detect the 105-kDa protein in the rat whole kidney until now. We investigated the detailed localization of the protein using water-restricted rat kidney segments, cortex, medulla, and papilla, on immunoblotting using an anti-rat 105-kDa protein IgG. As shown in Fig. 5, the protein was detected only in the soluble fraction of the renal medulla. We could not detect the 105-kDa protein in the insoluble fraction of renal medulla. None of the protein was detected in the soluble and insoluble fractions of renal cortex and renal papilla. The 105-kDa protein in the kidney is specifically located in the medulla, and the localization of the 105-kDa protein is the same as that ofOSP94. These results suggested that the 105-kDa protein is identical to OSP94 in partial amino-acid sequence and renal localization. Identification of APG-1/OSP94 and the 105-kDa protein To confirm the identity of APG-1/OSP94 and the 105-kDa protein, we studied the cross-reactivity of an antibody against APG-1 with the 105-kDa protein. As shown in Fig. 6, the purified 105-kDa protein was recognized by an antibody against the 105-kDa protein on immunoblotting. The 105-kDa protein was recognized by an antibody against APG-1 antibody, the same as an anti-105 kDa antibody. Based on these results, APG-1 and OSP94 are identical to the 105-kDa protein. Biochemical properties of the 105-kDa protein, APG-1, and OSP94 are shown in Table 1. Sequence homology between the 105-kDa protein (APG-1/OSP94) and HSP90 We analyzed the sequence homology between the 105-kDa protein (APG-1/OSP94) and mouse HSP90a (accession number P07901) and HSP90b (accession number P11499). There was a low homology (34.9 % match) between those Fig. 5. Interaction between the 105-kDa protein and OSP94. Control Wister rats or water-restricted rats (3 and 5 days) kidneys were dissected into renal cortex, medulla, and papilla. The soluble and insoluble fractions of these sections were electrophoresed on SDS- polyacrylamide gels (7% gel), which were stained with Coomassie Brilliant Blue (A), by immunoblotting with an antibody against the 105-kDa protein IgG (B). C, P, and M in panel A indicate renal cortex, renal papilla, and renal medulla, respectively. In panel A, 0, 3, and 5 indicate water-restriction time (days). Fig. 4. RT-HPLC fractionation of lysyl endopeptidase digests and the amino-acid sequence of the 105-kDa protein. A. Lysyl endopeptidase digests of the 105-kDa testis protein were separated by reverse phase chromatography on a C 18 column with a linear gradient of 0–60% acetonitrile in 0.1% TFA at a flow rate of 0.5 mL per minute. The purified peptides indicated in the panel (#40 and #70) were sequenced by a peptide sequencer. B. The two purified peptides of lysyl endo- peptidase digests (#40 and #70) were sequenced and compared with APG-1 and OSP94. Identical residues are denoted by a dash (–). Parentheses indicate the position of APG-1 and OSP94. Ó FEBS 2002 Biochemical properties of the 105-kDa protein (Eur. J. Biochem.)5 two proteins. Interestingly, we could detect a highly homologous region between APG-1/OSP94 (680–702) and mouse HSP90a or HSP90b (Fig. 7A). For the hydropathy profiles of APG-1/OSP94, the region showed hydrophilicity (Fig. 7B). Moreover, the domain may consist of a b-sheet and a b-turn on the secondary protein structure-prediction (Fig. 7C). Based on the analysis, the domain (670–700 from the N-terminal) may exist on the surface of APG-1/OSP94. On the contrary, we reported before that an antibody against bovine HSP90 recognizes mainly the N-terminal immunoreactive site of HSP90 (2–282 in the N-terminal region of human HSP90) [20]. The immunoreactive site is almost the same as the highly homologous site (264–284 from the N-terminal of HSP90) vs. those of APG-1/OSP94 (680–702 from the N-terminal). Based on these reasons, an anti-bovine HSP90 antibody may cross-react with the 105-kDa protein (APG-1/OSP94) the same as HSP90 in rat testis and brain. Expression time of 105-kDa protein in rat brain In order to clarify the localization of the 105-kDa protein in the brain, the rat brain was dissected into six sections and Table 1. Properties of the 105-kDa protein, APG-1, and OSP94. ND, not determined. Parameters 105-kDa protein APG-1 OSP94 Apparent molecular mass 105 kDa ND 105–110 kDa Chaperone activity + ND ND ATP-binding + + (putative) + (putative) Localization in organ Testis, brain Testis, brain Renal medulla renal medulla Encoded amino acids ND 838 838 Amino-acid sequence Partially same as APG-1and OSP94 Same as OSP94 Same as APG-1 Localization in testis Spermatozoa ND ND Binding protein p53 ND ND Fig. 6. Cross-reactivity of an antibody against APG-1 to the 105-kDa protein. Soluble fraction of rat brain and purified the 105-kDa protein were electrophoresed on SDS-polyacrylamide gels (10% gel), which were stained with Coomassie Brilliant Blue (A), by immunoblotting with an antibody against the 105-kDa protein IgG (B) or immuno- blotting with an antibody against APG-1 (C). Lane 1, purified 105-kDa protein; lane 2, soluble fraction of rat brain; lane 3, molecular standard proteins. Fig. 7. Secondary structure of the APG-1/OSP94 (the 105-kDa pro- tein). (A) Sequence homology between APG-1/OSP94 and mouse HSP94a (accession no. P07901) or HSP90b (accession no. P11499). The same amino acid and homologous amino acid are shown in red square and in yellow square, respectively. (B) Hydropathy profiles of APG-1/OSP94. In the panel, plus values indicate hydrophilicity of the amino-acid residues. (C) Secondary structure prediction of APG-1/ OSP94. Helix indicates blue loop, sheet structure indicates red zigzag, and turn structure indicates green line. Numbers in the panel indicate amino-acid residues of APG-1/OSP94. 6 M. Matsumori et al. (Eur. J. Biochem.) Ó FEBS 2002 the protein was detected on immunoblotting using an anti- (rat 105-kDa protein) IgG. The 105-kDa protein was detected in the soluble fraction. As shown in Fig. 1B, the 105-kDa protein was detected in all brain sections. The protein was detected in large amounts in the cerebellum and medulla oblongata compared to the other sections. On the contrary, the 105-kDa protein was detected as a faint protein band in the insoluble fractions (Fig. 8C). We investigated the expression time of the protein in the postnatal Wistar rat brain (0.5-, 1-, 2-, 3-, 4-, 5-, and 6-weeks-old) on immunoblotting. The 105-kDa protein could be detected at 3.5 s-day-old-in all sections of the rat brain. However, the induction pattern was different in each case. In the cerebral cortex and hippocampus, the expres- sion of the protein was strongly induced at 2-weeks-old (Fig. 8E). In the olfactory lobe, mid brain and medulla oblongata, the expressions of the 105-kDa protein were induced at 3 weeks old. In the cerebellum, the expression of the 105-kDa protein was induced at 4 weeks old (Fig. 8E). Thus, the expression time of the 105-kDa protein has been shown with slight differences in rat brain sections. Immunohistochemistry of the 105-kDa protein in rat brain In order to investigate the localization of the 105-kDa protein in the rat brain, we performed immunohistochem- ical studies using an anti-(rat 105-kDa protein) IgG. The localization of the 105-kDa protein in the rat brain is presented in Fig. 9. The protein is localized predominantly in the cytoplasm of nerve cells in the cerebral cortex, hippocampus, and cerebellum. Some neurons showed Fig. 8. Localization and expression time of the 105-kDa protein in rat brain. Rat brain (female 6 weeks) was dissected into six sections and soluble proteins were electrophoresed on SDS-polyacrylamide gels (10% gel), which were stained with Coomassie Brilliant Blue (A), by immunoblotting with an antibody against the 105-kDa protein IgG (B). Insoluble proteins were electrophoresed and processed immuno- blotting with an antibody against the 105-kDa protein IgG (C). Each section of the postnatal rat brain (3.5-day-old, 1-, 2-, 3-, 4-, 5-, and 6-week-old) were homogenized and the supernatants were electro- phoresed on SDS-polyacrylamide gels (10% gel), which were stained with Coomassie Brilliant Blue (D, rat cerebral cortex), by immuno- blotting with an antibody against the 105-kDa protein IgG (E). Fig. 9. Immunohistochemistry of the 105-kDa protein in rat brain. The sections of rat brain were stained with anti-rat 105-kDa protein IgG. Panels A and B, cerebral cortex; C and D, hippocampus; E and F, cerebellum. Bar indicates 50 lm. Ó FEBS 2002 Biochemical properties of the 105-kDa protein (Eur. J. Biochem.)7 nuclear staining. Cell bodies and proximal dendrites were intensely stained, whereas distal dendrites and axons were obscure. Glia cells were also reacted. Purkinje cells were the most intensely stained in the cerebral cortex. Detection of the 105-kDa protein in PC12 and PC12h cells The 105-kDa protein is localized in the rat brain, especially in nerve cells. We investigated the induction pattern of the protein in PC12 and PC12h (sub clone of PC12 cell) cells under different stress conditions. As shown in Fig. 10, HSP70 and HSP60 both were strongly induced under AzC treatment or heat treatment in both cells. We investigated the induction patterns of the 105-kDa protein under the same conditions. Although the protein was slightly induced under AzC treatment, the induction pattern of the protein under heat treatment was quite different from those of HSP70 and HSP60. The protein was slightly reduced under heat treatment in PC12 cells. Surprisingly, the protein was strongly reduced in PC12h cells. The same pattern was obtained from HSP90. Under these conditions, HSF-1 was activated in both cells (Fig. 10C). DISCUSSION We reported before that a 105-kDa testis and brain protein was cross-reacted with an antibody against bovine HSP90 [8]. On immunoblotting, the 105-kDa protein could not be detected in the liver, lung, spleen, kidney, ovarium and uterus, in contrast to the wide distribution of HSP90 [8]. The physicochemical properties of the 105-kDa protein were similar to those of HSP90, and the protein seems to be a cognate protein of HSP90 [8]. We produced a specific antibody against rat 105-kDa protein and the antibody was cross-reacted only with the protein in the rat brain. Recently, we have shown that the 105-kDa protein is induced by heat stress and is able to bind to p53 in a temperature-sensitive manner in the rat testis [10]. In the present study, we have characterized the biochemical properties of the 105-kDa protein. Mitochondrial CS was chosen as a model substrate in the chaperone activity. CS aggregates and is inactived rapidly upon incubation at 43 °C [21]. The 105-kDa protein inhibited CS aggregation. The protein binds transiently to unfolding intermediates of the thermal unfolding of CS. Upon release from the 105-kDa protein, the intermediates are able to refold rapidly to the native state. Thus, the 105-kDa protein stabilizes the native CS. Chaperone activity of the 105-kDa protein was thesameasthatofHSP90[7]. Many molecular chaperones including HSP70 show ATPase activity or bind to ATP–Sepharose [13]. For ATP binding of the 105-kDa protein, ATP–Sepharose column chromatography has been performed. On immunoblotting, the protein was detected in all eluted fractions from the column. These results suggest that the protein is an ATP- binding protein the same as the other molecular chaperones and may have an ATP-binding sequence. It has been shown that the ATP-binding consensus sequence is divided into two short elements termed type A, the putative triphosphate binding sequence, and type B, on adenine-binding sequence [22,23]. Type A is A/GXXXXGKT/SXXXXXXI/V. On the contrary, type B is H/RKK (5)7) hXhhD/E, where h stands for a hydrophobic residue. We searched for ATP- binding proteins with a molecular mass of about 100 kDa in the data base. Two interesting proteins, APG-1 and OSP94, have been found. An HSP110-related gene, APG-1, has been isolated from the mouse testis cDNA library [11]. APG-1 was abundantly expressed in the testis, and a lower level of expression was seen in the brain on Northern blot analysis [11]. On the contrary, OSP94 cDNA, a member of the HSP110/SSE family, has been cloned from another group [12]. Renal inner medullary OSP94 mRNA expression was increased in water restricted mice. Both APG-1 and OSP94 cDNA encodes an 838-amino-acid residues protein, and the sequence is the same in each case. Both genes show the putative ATP-binding sequence in the amino terminal. Therefore, these two genes, APG-1 and OSP94, are the same. The in vitro translated OSP94 product migrated as the 105–110-kDa protein on SDS/PAGE [12]. The specific localization of APG-1/OSP94 in mouse organs and their Fig. 10. Induction pattern of the 105-kDa protein in PC12 and PC12h cells. PC12 and PC12h cells were stressed with 5 m M AzC for 6 h or stressed by the increasing of temperature at 43 °Cfor30minasdes- cribed under Materials and methods. Cells were homogenized with SDS-sample buffer and the samples were electrophoresed on SDS- polyacrylamide gels (7% gel), which were stained with Coomassie Brilliant Blue (A), by immunoblotting with an antibodies against HSP60, HSP70, HSP90, and rat 105-kDa protein (B) or immuno- blotting with an antibody against HSF-1 (C). 8 M. Matsumori et al. (Eur. J. Biochem.) Ó FEBS 2002 apparent molecular masses are the same as the 105-kDa protein. To analyze the partial amino-acid sequence of the 105-kDa protein, lysyl endopeptidase digested fragments were separated on HPLC, and the partial amino-acid sequence was determined. The 24 amino-acid residues obtained from two different peptides were the same as the deduced amino-acid sequence of APG-1 and OSP94. An important question regarding localization of the 105-kDa protein in the kidney has remained unanswered. Although the protein could not be detected in the whole kidney sample on immunoblotting, the protein could be clearly detected only in the medulla of normal and 3- or 5-day water- restricted rat kidney. It has been shown that OSP94 mRNA expression was increased in the renal medulla during water restriction. In the present study, we could not detect the increasing expression of the protein in the water-restricted rat kidneys. This may reflect the difference in detection systems between Northern blotting and immunoblotting. An anti-APG-1 cross-reacted with the 105-kDa protein on immunoblotting. The former detects mRNA and the latter detect the protein. These results allow us to conclude that APG-1 and OSP94 are identical to the 105-kDa protein. The most important question has remained unanswered. Why does an antibody against bovine HSP90 cross-react with the 105-kDa protein (APG-1/OSP94)? As shown in Fig. 7, there were highly homologous regions between the immunoreactive sites of HSP90 (2–282 in the N-terminal region of human HSP90) and the 105-kDa protein (APG-1/ OSP94). An anti-HSP90 antibody may recognize the 105-kDa protein the same as HSP90. In 1997, APG-1 and OSP94 cDNA has been isolated independently [11,12]. However, the biochemical properties of the two proteins have not yet been fully understood until today. In 1990, we purified and reported some biochemical properties of the 105-kDa protein as a novel HSP90-related protein [8]. Now, we have shown the identity of APG-1 and OSP94 relative to the 105-kDa protein and that the 105-kDa protein plays a role as a molecular chaperone. We have reported the physiological functions of the protein in the rat testis [9,10]. The protein could bind to p53 in a temperature-dependent manner in the cyto- plasm of the germ cells. The 105-kDa protein may contribute to the stabilization of p53 and prevent the potential induction of apoptosis by p53. On the contrary, localization and expression time of the protein in the brain are not yet known. Immunoblotting using an anti-105 kDa protein IgG revealed that the 105-kDa protein was constitutively expressed in all sections of the rat brain. In the present study, we showed that the 105- kDa protein was localized in the cytoplasm of the nerve cells and/or glia of hippocampus, cerebral cortex, and cerebellum of the rat brain. The 105-kDa protein may play an important role in brain throughout postnatal period. PC12h cells had been established as one of the sub clones of PC12 cells [24,25]. The cell was demonstrated to have nerve growth factor (NGF)-responsive tyrosine hydroxylase (TH) activity. In the present investigation, HSP70 and HSP60 were strongly induced by AzC, an amino-acid analog (proline-synthesis inhibitor), or heat treatment of PC12h cells. HSF-1 was activated under these stress-conditions. Surprisingly, both 105-kDa protein and HSP90 both were reduced under the heat stress conditions in spite of the remarkable induction of HSP70 and HSP60 under the same conditions. Although APG-1, OSP94, and the 105-kDa protein have homology to HSP110 and HSP70 in their primary amino-acid sequences, the induction pattern under the heat stress conditions in PC12h cells were quite different from those of HSP70 and HSP60. The reason why the 105-kDa protein and HSP90 were reduced under the stressed conditions of PC12h cells are not yet known at present. The 105-kDa protein and HSP90 induced under the heat treatment might be quickly digested by protease in PC12h cells. We investigated the induction pattern of HSP90 and the 105-kDa protein under the same conditions in the presence of some kinds of protease inhibitors. However, the induction patterns of the 105-kDa protein and HSP90 were the same as those in the absence of protease inhibitors (data not shown). Thus, the different induction patterns of the 105-kDa protein and HSP90 from those of the other molecular chaperones in PC12h cells might be dependent on the difference in transcriptional mechanisms. It has been shown that the regulation of heat induction of the APG-1 transcript cannot be explained by the HSF1 activation alone and that some other mechanisms are responsible for the differential induction of HSP70 and APG-1 [11]. The transcriptional mechanisms of the 105-kDa protein may be different from those of the other molecular chaperones including HSP70 in the PC12h cells. By inves- tigating the biochemical properties of the 105-kDa protein in PC12h cells, we may be able to understand the physiological functions of the protein in nerve cells. We have already started research to understand the biochemical properties of the protein using over-expression of the protein or HSP90 in PC12h cells. ACKNOWLEDGEMENTS This work was supported in part by a Grant-in-Aid for Scientific Research on Priority Areas (C) (Advanced Brain Science Project: 12210033 to H. I.) and by a Grant-in-Aid for Scientific Research on Priority Areas (Molecular Chaperone: 11153201 to H. I.) and C2 (12670105 to H. I., 14571011-00 to A. K., and 14570442 to M. O.) from the Ministry of Education, Culture, Sports, Science and Technology, Japan. REFERENCES 1. Lindquist, S. (1986) The heat shock response. Annu. Rev. Biochem. 55, 1151–1191. 2. Catelli, M.G., Binart, N., Jung-Testas, I., Renoir, J.M., Baulieu, E.E., Feramisco, J.R. & Welch, W.J. (1985) The common 90-kd protein component of non-transformed Ô8SÕ steroid receptors is a heat-shock protein. EMBO J. 4, 3131–3135. 3. Rutherford,S.&Lindquist,S.(1998)Hsp90asacapacitorfor morphological evolution. Nature 396, 336–342. 4. Weich, H., Buchner, J., Zimmermann, R. & Jakob, U. (1992) Hsp90 as a capacitor for morphological evolution. Nature 358, 169–170. 5. Young, J.C., Schneider, C. & Hartl, F.U. (1997) In vitro evidence that hsp90 contains two independent chaperone sites. FEBS Lett. 418, 139–143. 6. Scheibel, T., Weikl, T. & Buchner, J. (1999) Two chaperone sites in Hsp90 differing in substrate specificity and ATP dependence. Proc. Natl Acad. Sci. USA 95, 1495–1499. Ó FEBS 2002 Biochemical properties of the 105-kDa protein (Eur. J. Biochem.)9 7. Itoh, H., Ogura, M., Komatsuda, A., Wakui, H., Miura, A.B. & Tashima, Y. (1998) A novel chaperone-activity-reducing mechan- ism of the 90-kDa molecular chaperone HSP90. Biochem. J. 343, 697–703. 8. Itoh, H. & Tashima, Y. (1990) A novel testis-specific 105-kDa protein related to the 90-kDa heat-shock protein. Eur. J. Biochem. 193, 429–435. 9. Itoh, H. & Tashima, Y. (1991) Different expression time of the 105-kDa protein and 90-kDa heat-shock protein in rat testis. FEBS Lett. 289, 110–112. 10. Kumagai, J., Fukuda, J., Kodama, H., Murata, M., Kawamura, K., Itoh, H. & Tanaka, T. (2000) Germ cell-specific heat shock protein 105 binds to p53 in a temperature-sensitive manner in rat testis. Eur. J. Biochem. 267, 3073–3078. 11. Kaneko, Y., Nishiyama, H., Nonoguchi, K., Higashitsuji, H. & Kishishita, M. & Fujita, J. (1997) A novel hsp110-related gene, apg-1, that is abundantly expressed in the testis responds to a low temperature heat shock rather than the traditional elevated tem- peratures. J. Biol. Chem. 272, 2640–2645. 12. Kojima, R., Randall, J., Brenner, B.M. & Gullans, S.R. (1996) Osmotic stress protein94 (Osp94). A new member of the Hsp110/ SSE gene subfamily. J. Biol. Chem. 271, 12327–12332. 13. Itoh, H. & Tashima, Y. (1993) Localization of HSP90 in rat brain. Int. J. Biochem. 25, 69–77. 14. Itoh, H., Kobayashi, R., Wakui, H., Komatsuda, A., Ohtani, H., Miura, B.A., Otaka, M., Masamune, O., Andoh, H., Koyama, K., Sato, Y. & Tashima, Y. (1995) Mammalian 60-kDa stress protein (chaperonin homolog). Identification, biochemical properties, and localization. J. Biol. Chem. 270, 13429–13454. 15. Itoh,H.,Komatsuda,A.,Wakui,H.,Miura,B.A.&Tashima,Y. (1999) Mammalian HSP60 is a major target for an immunosuppressant mizoribine. J. Biol. Chem. 274, 35147–35151. 16. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. 17. Towbin,H.,Staechelin,T.&Gordon,J.(1979)Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl Acad. Sci. USA 76, 4350–4354. 18. Soti, C., Racz, A. & Csermely, P. (2002) A nucleotide-dependent molecular switch controls ATP binding at the C-terminal domain of Hsp90. N-terminal nucleotide binding unmasks a C-terminal binding pocket. J. Biol. Chem. 277, 7066–7075. 19. Yasuda, K., Nakai, A., Hatayama, T. & Nagata, K. (1995) Cloning and expression of murine high molecular mass heat shock proteins, HSP105. J. Biol. Chem. 270, 29718–29723. 20. Itoh, H. & Tashima, Y. (1993) Domain structure of the 90-kDa stress protein, heparin- and antibody-binding domain. Int. J. Biochem. 25, 157–161. 21. Jacob, U., Lille, H., Meyer, L. & Buchnner, J. (1995) Transient interaction of Hsp90 with early unfolding intermediates of citrate synthase. Implications for heat shock in vivo. J. Biol. Chem. 270, 7288–7294. 22. Walker, J.E., Saraste, M., Runswick, M.J. & Gay, N.J. (1982) Distantly related sequences in the alpha- and beta-subunits of ATP synthase, myosin, kinases and other ATP-requiring enzymes and a common nucleotide binding fold. EMBO J. 1, 945–951. 23. Chin, D.T., Goff, S.A., Webster, T., Smith, T. & Goldberg, A.L. (1988) Sequence of the lon gene in Escherichia coli. A heat-shock gene which encodes the ATP-dependent protease La. J. Biol. Chem. 263, 11718–11728. 24. Hatanaka, H. (1981) Nerve growth factor-mediated stimulation of tyrosine hydroxylase activity in a clonal rat pheochromocytoma cell line. Brain Res. 222, 225–233. 25. Ikenaka, K., Nakahira, K., Takayama, C., Wada, K., Hatanaka, H. & Mikoshiba, K. (1990) Nerve growth factor rapidly induces expression of the 68-kDa neurofilament gene by posttranscrip- tional modification in PC12h-R. cells. J. Biol. Chem. 265, 19782–19785. 10 M. Matsumori et al. (Eur. J. Biochem.) Ó FEBS 2002 . renal cortex and renal papilla. The 105-kDa protein in the kidney is specifically located in the medulla, and the localization of the 105-kDa protein is the same. biochemical properties of the 105-kDa protein. In the present study, we discuss the biochemical properties of the 105-kDa protein in the rat brain and the interaction